++

generated_alpha/2025-12-15/104029.txt

@@ -0,0 +1,221 @@
+group_mean(if_else(ts_rank(volume, 30) > 0.7, ts_delta(close, 10), ts_delta(close, 30)), 1, industry)
+
+multiply(ts_corr(returns, ts_delay(returns, 5), 20), group_mean(ts_delta(close, 5), 1, sector))
+
+if_else(ts_std_dev(returns, 60) > ts_mean(ts_std_dev(returns, 60), 120), ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(ts_delta(close, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(reverse(ts_rank(volume / ts_mean(volume, 20), 10)), ts_delta(close, 20))
+
+multiply(ts_regression(close, ts_step(1), 30, 0, 1), group_zscore(volume, industry))
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_mean(returns, 20))
+
+group_mean(ts_decay_linear(returns, 10), 1, bucket(ts_rank(volume, 30), range="0,3,0.4"))
+
+multiply(ts_corr(group_mean(returns, 1, sector), returns, 30), ts_zscore(close, 60))
+
+if_else(bucket(rank(volume), range="0,3,0.4") == 0, ts_delta(close, 10), ts_delta(close, 30))
+
+multiply(ts_delta(close, 5), reverse(ts_rank(ts_delta(volume, 5), 10)))
+
+group_mean(if_else(ts_mean(returns, 20) > 0, ts_delta(close, 10), reverse(ts_delta(close, 10))), 1, industry)
+
+ts_delta(close, if_else(ts_std_dev(returns, 20) > 0.02, 5, 20))
+
+multiply(ts_rank(ts_delta(close, 5), 30), group_rank(volume, sector))
+
+group_mean(ts_corr(returns, ts_mean(volume, 5), 20), 1, bucket(rank(close), range="0,3,0.4"))
+
+if_else(ts_rank(ts_mean(returns, 60), 120) > 0.5, ts_delta(close, 10), ts_delta(close, 30))
+
+multiply(group_zscore(ts_delta(close, 5), industry), ts_rank(volume, 20))
+
+group_mean(ts_delta(close, 10) / ts_std_dev(returns, 20), 1, sector)
+
+multiply(ts_regression(close, ts_step(1), 20, 0, 1), if_else(rank(volume) > 0.8, 1, 0.5))
+
+ts_mean(returns, if_else(ts_rank(volume, 30) > 0.8, 10, 30))
+
+multiply(group_rank(ts_delta(close, 5), industry), reverse(ts_rank(ts_delta(volume, 5), 10)))
+
+group_mean(if_else(ts_std_dev(returns, 10) > ts_mean(ts_std_dev(returns, 10), 30), ts_delta(close, 5), ts_delta(close, 20)), 1, industry)
+
+multiply(ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 20), group_mean(returns, 1, sector))
+
+if_else(ts_rank(ts_mean(volume, 20), 60) > 0.7, ts_delta(close, 10), ts_mean(returns, 20))
+
+group_mean(ts_zscore(ts_delta(close, 20), 30), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_delta(close, 10), reverse(ts_rank(volume / ts_mean(volume, 30), 20)))
+
+ts_delta(close, 5) / ts_std_dev(returns, if_else(ts_rank(volume, 30) > 0.7, 10, 30))
+
+multiply(group_mean(ts_regression(close, ts_step(1), 20, 0, 1), 1, industry), ts_rank(returns, 10))
+
+if_else(bucket(ts_rank(volume, 30), range="0,3,0.4") == 0, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(multiply(ts_delta(close, 10), ts_zscore(volume, 20)), 1, sector)
+
+multiply(ts_rank(ts_decay_linear(returns, 5), 20), group_zscore(close, industry))
+
+ts_mean(returns, if_else(ts_std_dev(returns, 20) > 0.015, 5, 20))
+
+group_mean(if_else(ts_delta(close, 1) > 0, ts_delta(close, 10), reverse(ts_delta(close, 10))), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_corr(ts_mean(returns, 5), ts_mean(volume, 5), 30), ts_delta(close, 10))
+
+if_else(ts_rank(ts_std_dev(returns, 30), 60) > 0.6, ts_delta(close, 5), ts_regression(close, ts_step(1), 30, 0, 1))
+
+group_mean(ts_delta(close, 20) * ts_rank(volume, 20), 1, industry)
+
+multiply(reverse(ts_rank(ts_delta(volume, 1), 10)), group_mean(ts_delta(close, 5), 1, sector))
+
+ts_delta(close, if_else(ts_mean(returns, 20) > 0, 10, 30))
+
+group_mean(ts_zscore(ts_delta(close, 5), 20), 1, bucket(ts_rank(volume, 30), range="0,3,0.4"))
+
+multiply(ts_regression(close, ts_step(1), 10, 0, 1), if_else(rank(volume) > 0.9, 2, 1))
+
+if_else(ts_rank(ts_mean(volume, 30), 90) > 0.8, ts_delta(close, 5), ts_mean(returns, 30))
+
+group_mean(multiply(ts_delta(close, 10), reverse(ts_rank(volume, 20))), 1, industry)
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 10) / ts_std_dev(returns, 20)
+
+multiply(group_rank(ts_mean(returns, 20), sector), ts_zscore(ts_delta(close, 5), 30))
+
+if_else(bucket(rank(ts_delta(volume, 1)), range="0,3,0.4") == 0, ts_delta(close, 10), ts_delta(close, 30))
+
+group_mean(ts_delta(close, 5) / ts_mean(abs(returns), 20), 1, sector)
+
+multiply(ts_rank(ts_decay_linear(returns, 10), 30), group_mean(volume, 1, industry))
+
+ts_mean(returns, if_else(ts_rank(ts_std_dev(returns, 10), 30) > 0.7, 5, 20))
+
+group_mean(if_else(ts_delta(volume, 1) > 0, ts_delta(close, 10), ts_delta(close, 20)), 1, bucket(rank(close), range="0,3,0.4"))
+
+multiply(ts_corr(group_mean(ts_delta(close, 1), 1, sector), returns, 20), ts_regression(close, ts_step(1), 20, 0, 1))
+
+if_else(ts_rank(ts_mean(returns, 30), 60) > 0.5, ts_delta(close, 10), reverse(ts_delta(close, 10)))
+
+group_mean(ts_zscore(ts_delta(close, 20), 40), 1, industry)
+
+multiply(reverse(ts_rank(volume / ts_delay(volume, 1), 10)), ts_delta(close, 10))
+
+ts_delta(close, 5) * if_else(ts_rank(volume, 30) > 0.7, 1.5, 1)
+
+group_mean(multiply(ts_regression(close, ts_step(1), 30, 0, 1), ts_rank(volume, 20)), 1, sector)
+
+if_else(ts_std_dev(returns, 20) > ts_mean(ts_std_dev(returns, 20), 60), ts_delta(close, 5), ts_mean(returns, 20))
+
+multiply(group_zscore(ts_delta(close, 10), industry), reverse(ts_rank(ts_delta(volume, 5), 15)))
+
+ts_mean(returns, if_else(bucket(rank(volume), range="0,3,0.4") == 0, 10, 30))
+
+group_mean(ts_delta(close, 15) / ts_std_dev(returns, 30), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_corr(ts_mean(returns, 10), ts_mean(volume, 10), 20), group_rank(close, sector))
+
+if_else(ts_rank(ts_delta(volume, 1), 30) > 0.8, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(if_else(ts_mean(returns, 10) > 0, ts_delta(close, 10), ts_delta(close, 30)), 1, industry)
+
+multiply(ts_rank(ts_delta(close, 5), 20), group_zscore(volume, sector))
+
+ts_delta(close, if_else(ts_rank(ts_std_dev(returns, 10), 30) > 0.6, 5, 20))
+
+group_mean(ts_zscore(ts_delta(close, 10), 25), 1, bucket(ts_rank(volume, 20), range="0,3,0.4"))
+
+multiply(reverse(ts_rank(ts_mean(volume, 5), 10)), group_mean(ts_delta(close, 5), 1, industry))
+
+if_else(ts_rank(ts_mean(returns, 20), 40) > 0.5, ts_delta(close, 10), reverse(ts_delta(close, 10)))
+
+multiply(ts_corr(returns, group_mean(returns, 1, sector), 30), ts_regression(close, ts_step(1), 25, 0, 1))
+
+group_mean(ts_delta(close, 20) * if_else(rank(volume) > 0.8, 1.2, 0.8), 1, sector)
+
+ts_mean(returns, if_else(ts_std_dev(returns, 15) > 0.02, 5, 25))
+
+multiply(group_rank(ts_mean(returns, 15), industry), reverse(ts_rank(volume / ts_mean(volume, 30), 20)))
+
+if_else(bucket(ts_rank(ts_delta(volume, 1), 20), range="0,3,0.4") == 0, ts_delta(close, 10), ts_mean(returns, 20))
+
+group_mean(ts_delta(close, 10) / ts_mean(abs(returns), 30), 1, bucket(rank(close), range="0,3,0.4"))
+
+multiply(ts_rank(ts_decay_linear(returns, 8), 25), group_mean(ts_zscore(volume, 20), 1, industry))
+
+ts_delta(close, 5) * ts_rank(volume, 10) / ts_std_dev(returns, 20)
+
+group_mean(if_else(ts_delta(volume, 1) > ts_mean(ts_delta(volume, 1), 10), ts_delta(close, 5), ts_delta(close, 15)), 1, sector)
+
+multiply(ts_corr(ts_delta(close, 3), ts_delta(volume, 3), 15), group_zscore(returns, industry))
+
+if_else(ts_rank(ts_std_dev(returns, 25), 50) > 0.7, ts_delta(close, 5), ts_regression(close, ts_step(1), 30, 0, 1))
+
+group_mean(ts_zscore(ts_delta(close, 12), 24), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(reverse(ts_rank(ts_delta(volume, 2), 12)), ts_delta(close, 12))
+
+ts_mean(returns, if_else(ts_rank(volume, 30) > 0.9, 8, 22))
+
+group_mean(multiply(ts_regression(close, ts_step(1), 18, 0, 1), ts_rank(ts_delta(volume, 1), 15)), 1, industry)
+
+if_else(ts_rank(ts_mean(returns, 18), 36) > 0.5, ts_delta(close, 8), ts_delta(close, 22))
+
+multiply(ts_corr(group_mean(returns, 1, sector), ts_delay(returns, 1), 20), group_rank(ts_delta(close, 5), industry))
+
+group_mean(ts_delta(close, 15) / if_else(ts_std_dev(returns, 20) > 0.018, ts_std_dev(returns, 20), 0.01), 1, sector)
+
+multiply(ts_rank(ts_delta(close, 6), 18), reverse(group_zscore(volume, industry)))
+
+ts_delta(close, if_else(bucket(rank(ts_mean(volume, 20)), range="0,3,0.4") == 0, 6, 18))
+
+group_mean(if_else(ts_mean(returns, 15) > group_mean(ts_mean(returns, 15), 1, industry), ts_delta(close, 10), ts_delta(close, 20)), 1, sector)
+
+multiply(ts_decay_linear(returns, 10), group_mean(ts_rank(volume, 20), 1, industry))
+
+if_else(ts_rank(ts_std_dev(returns, 22), 44) > 0.65, ts_delta(close, 7), ts_mean(returns, 22))
+
+group_mean(ts_zscore(ts_delta(close, 8), 16), 1, bucket(ts_rank(returns, 20), range="0,3,0.4"))
+
+multiply(reverse(ts_rank(volume / ts_delay(volume, 5), 15)), group_mean(ts_regression(close, ts_step(1), 20, 0, 1), 1, sector))
+
+ts_delta(close, 10) * if_else(ts_rank(ts_delta(volume, 5), 30) > 0.8, 1.3, 0.7)
+
+group_mean(multiply(ts_corr(returns, ts_mean(volume, 10), 25), ts_delta(close, 9)), 1, industry)
+
+if_else(ts_rank(ts_mean(volume, 25), 50) > 0.75, ts_delta(close, 6), ts_delta(close, 18))
+
+multiply(group_rank(ts_zscore(close, 30), sector), ts_delta(close, 11))
+
+group_mean(ts_delta(close, 14) / ts_std_dev(returns, 28), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_rank(ts_mean(returns, 12), 24), reverse(group_mean(ts_rank(volume, 15), 1, industry)))
+
+ts_mean(returns, if_else(ts_std_dev(returns, 18) > ts_mean(ts_std_dev(returns, 18), 36), 7, 21))
+
+group_mean(if_else(ts_delta(close, 1) > ts_mean(ts_delta(close, 1), 10), ts_delta(close, 8), reverse(ts_delta(close, 8))), 1, sector)
+
+multiply(ts_corr(ts_delta(close, 4), ts_delta(volume, 4), 18), group_zscore(ts_delta(close, 4), industry))
+
+if_else(bucket(ts_rank(ts_std_dev(returns, 15), 30), range="0,3,0.4") == 0, ts_delta(close, 5), ts_delta(close, 25))
+
+group_mean(ts_delta(close, 16) * ts_rank(ts_delta(volume, 2), 20), 1, industry)
+
+multiply(reverse(ts_rank(ts_mean(volume, 8), 16)), ts_regression(close, ts_step(1), 22, 0, 1))
+
+ts_delta(close, if_else(ts_rank(returns, 30) > 0.7, 9, 27))
+
+group_mean(ts_zscore(ts_delta(close, 7), 14), 1, bucket(rank(ts_delta(volume, 1)), range="0,3,0.4"))
+
+multiply(ts_mean(returns, 15), if_else(group_rank(volume, sector) > 0.8, 1.5, 0.5))
+
+if_else(ts_rank(ts_mean(returns, 28), 56) > 0.55, ts_delta(close, 10), reverse(ts_delta(close, 10)))
+
+group_mean(multiply(ts_corr(ts_mean(returns, 8), group_mean(ts_mean(returns, 8), 1, sector), 20), ts_delta(close, 10)), 1, industry)
+
+ts_delta(close, 5) / if_else(ts_rank(volume, 30) > 0.85, ts_std_dev(returns, 10), ts_std_dev(returns, 30))
+
+multiply(group_zscore(ts_regression(close, ts_step(1), 25, 0, 1), sector), reverse(ts_rank(ts_delta(volume, 3), 18)))

generated_alpha/2025-12-15/104255.txt

@@ -0,0 +1,279 @@
+multiply(ts_delta(close,10), group_mean(ts_delta(close,10), 1, bucket(rank(volume), range="0,3,0.4")))
+
+if_else(ts_rank(ts_std_dev(returns,20), 60) > 0.7, ts_delta(close,5), ts_delta(close,20))
+
+ts_corr(ts_delta(close,5), ts_delta(volume,5), 30)
+
+ts_regression(close, ts_step(1), 30, 0, 1)
+
+ts_decay_linear(ts_delta(close,5), 20)
+
+multiply(ts_mean(returns,20), ts_std_dev(returns,20))
+
+ts_delta(ts_delta(close,10), 5)
+
+group_mean(ts_delta(close,20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(rank(volume) > 0.7, ts_delta(close,20), ts_delta(close,10))
+
+ts_rank(ts_mean(returns,20), 60) > 0.5
+
+ts_zscore(ts_delta(close,30), 60)
+
+multiply(ts_delta(close,60), ts_std_dev(returns,20))
+
+ts_arg_max(returns, 30)
+
+ts_arg_min(returns, 30)
+
+ts_av_diff(returns, 20)
+
+ts_count_nans(returns, 30)
+
+ts_covariance(returns, volume, 20)
+
+ts_backfill(ts_delta(close,5), 30)
+
+ts_delay(close, 5)
+
+ts_mean(ts_delta(close,5), 30)
+
+ts_product(returns, 10)
+
+ts_quantile(returns, 20)
+
+ts_scale(close, 60)
+
+ts_sum(returns, 20)
+
+ts_zscore(volume, 30)
+
+normalize(ts_delta(close,10), true)
+
+quantile(ts_delta(close,10))
+
+rank(ts_delta(close,20))
+
+scale(ts_delta(close,20))
+
+winsorize(ts_delta(close,20), 3)
+
+zscore(ts_delta(close,20))
+
+vec_avg(ts_delta(close,20))
+
+vec_sum(ts_delta(close,20))
+
+bucket(rank(volume), range="0,3,0.4")
+
+group_backfill(ts_delta(close,10), industry, 30)
+
+group_mean(ts_delta(close,10), 1, industry)
+
+group_neutralize(ts_delta(close,10), sector)
+
+group_rank(ts_delta(close,10), industry)
+
+group_scale(ts_delta(close,10), sector)
+
+group_zscore(ts_delta(close,10), sector)
+
+multiply(ts_delta(close,20), if_else(ts_rank(ts_std_dev(returns,20), 60) > 0.7, 1, -1))
+
+divide(ts_delta(close,10), ts_std_dev(returns,20))
+
+reverse(ts_delta(close,20))
+
+sign(ts_delta(close,10))
+
+signed_power(ts_delta(close,10), 2)
+
+sqrt(ts_mean(returns,20))
+
+subtract(ts_delta(close,20), ts_delta(close,10))
+
+add(ts_delta(close,20), ts_delta(close,10))
+
+max(ts_delta(close,20), ts_delta(close,10))
+
+min(ts_delta(close,20), ts_delta(close,10))
+
+power(ts_delta(close,10), 2)
+
+and(ts_rank(ts_std_dev(returns,20), 60) > 0.7, ts_delta(close,10) > 0)
+
+or(ts_rank(ts_std_dev(returns,20), 60) > 0.7, ts_delta(close,10) > 0)
+
+not(ts_rank(ts_std_dev(returns,20), 60) > 0.7)
+
+is_nan(ts_delta(close,10))
+
+if_else(ts_rank(ts_std_dev(returns,20), 60) > 0.7, ts_delta(close,5), ts_delta(close,60))
+
+multiply(ts_delta(close,10), ts_corr(ts_delta(close,5), ts_delta(volume,5), 20))
+
+ts_regression(close, ts_step(1), 60, 0, 1)
+
+ts_decay_linear(returns, 30)
+
+ts_mean(ts_delta(close,1), 20)
+
+ts_std_dev(ts_delta(close,5), 20)
+
+ts_sum(ts_delta(close,5), 30)
+
+ts_zscore(ts_delta(close,10), 30)
+
+group_mean(ts_delta(close,10), 1, sector)
+
+group_rank(ts_delta(close,10), sector)
+
+group_scale(ts_delta(close,10), sector)
+
+group_zscore(ts_delta(close,10), sector)
+
+bucket(rank(volume), range="0,5,0.2")
+
+ts_backfill(ts_delta(close,10), 60)
+
+ts_delay(returns, 1)
+
+ts_mean(returns, 10)
+
+ts_product(ts_delta(close,5), 5)
+
+ts_quantile(ts_delta(close,10), 30)
+
+ts_scale(ts_delta(close,10), 30)
+
+ts_sum(ts_delta(close,5), 10)
+
+ts_zscore(ts_delta(close,5), 10)
+
+normalize(ts_delta(close,5), false)
+
+quantile(ts_delta(close,5))
+
+rank(ts_delta(close,5))
+
+scale(ts_delta(close,5))
+
+winsorize(ts_delta(close,5), 4)
+
+zscore(ts_delta(close,5))
+
+vec_avg(ts_delta(close,5))
+
+vec_sum(ts_delta(close,5))
+
+group_backfill(ts_delta(close,5), industry, 20)
+
+group_mean(ts_delta(close,5), 1, industry)
+
+group_neutralize(ts_delta(close,5), industry)
+
+group_rank(ts_delta(close,5), industry)
+
+group_scale(ts_delta(close,5), industry)
+
+group_zscore(ts_delta(close,5), industry)
+
+multiply(ts_delta(close,5), ts_corr(ts_delta(close,5), ts_delta(volume,5), 10))
+
+divide(ts_delta(close,5), ts_std_dev(returns,10))
+
+reverse(ts_delta(close,5))
+
+sign(ts_delta(close,5))
+
+signed_power(ts_delta(close,5), 2)
+
+sqrt(ts_mean(returns,10))
+
+subtract(ts_delta(close,5), ts_delta(close,1))
+
+add(ts_delta(close,5), ts_delta(close,1))
+
+max(ts_delta(close,5), ts_delta(close,1))
+
+min(ts_delta(close,5), ts_delta(close,1))
+
+power(ts_delta(close,5), 2)
+
+and(ts_rank(ts_std_dev(returns,10), 30) > 0.7, ts_delta(close,5) > 0)
+
+or(ts_rank(ts_std_dev(returns,10), 30) > 0.7, ts_delta(close,5) > 0)
+
+not(ts_rank(ts_std_dev(returns,10), 30) > 0.7)
+
+is_nan(ts_delta(close,5))
+
+if_else(ts_rank(ts_std_dev(returns,10), 30) > 0.7, ts_delta(close,5), ts_delta(close,20))
+
+multiply(ts_delta(close,5), ts_corr(ts_delta(close,5), ts_delta(volume,5), 30))
+
+ts_regression(close, ts_step(1), 20, 0, 1)
+
+ts_decay_linear(returns, 10)
+
+ts_mean(ts_delta(close,1), 5)
+
+ts_std_dev(ts_delta(close,5), 10)
+
+ts_sum(ts_delta(close,5), 5)
+
+ts_zscore(ts_delta(close,5), 5)
+
+group_mean(ts_delta(close,5), 1, sector)
+
+group_rank(ts_delta(close,5), sector)
+
+group_scale(ts_delta(close,5), sector)
+
+group_zscore(ts_delta(close,5), sector)
+
+bucket(rank(volume), range="0,2,0.5")
+
+ts_backfill(ts_delta(close,5), 20)
+
+ts_delay(returns, 5)
+
+ts_mean(returns, 5)
+
+ts_product(ts_delta(close,5), 3)
+
+ts_quantile(ts_delta(close,5), 10)
+
+ts_scale(ts_delta(close,5), 10)
+
+ts_sum(ts_delta(close,5), 3)
+
+ts_zscore(ts_delta(close,5), 3)
+
+normalize(ts_delta(close,5), false)
+
+quantile(ts_delta(close,5))
+
+rank(ts_delta(close,5))
+
+scale(ts_delta(close,5))
+
+winsorize(ts_delta(close,5), 3)
+
+zscore(ts_delta(close,5))
+
+vec_avg(ts_delta(close,5))
+
+vec_sum(ts_delta(close,5))
+
+group_backfill(ts_delta(close,5), industry, 10)
+
+group_mean(ts_delta(close,5), 1, industry)
+
+group_neutralize(ts_delta(close,5), industry)
+
+group_rank(ts_delta(close,5), industry)
+
+group_scale(ts_delta(close,5), industry)
+
+group_zscore(ts_delta(close,5), industry)

generated_alpha/2025-12-15/104334.txt

@@ -0,0 +1,228 @@
+
+group_mean(ts_delta(close, 5), 1, bucket(rank(volume), range="0,3,0.4"))
+
+group_mean(ts_delta(close, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+group_mean(ts_regression(close, ts_step(1), 30, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_std_dev(returns, 20) > 0.02, ts_delta(close, 5), ts_delta(close, 20))
+
+if_else(ts_rank(ts_std_dev(returns, 60), 120) > 0.7, ts_delta(close, 10), ts_delta(close, 60))
+
+multiply(ts_corr(group_mean(returns, 1, industry), group_mean(returns, 1, sector), 20), ts_delta(close, 30))
+
+multiply(reverse(ts_rank(volume/ts_mean(volume, 20), 10)), ts_delta(close, 20))
+
+ts_delta(close, 5) / ts_std_dev(returns, 20)
+
+divide(ts_delta(close, 30), ts_mean(abs(returns), 30))
+
+if_else(bucket(rank(volume), range="0,3,0.4") == 0, ts_mean(returns, 10), ts_delta(close, 5))
+
+group_zscore(ts_delta(close, 10), bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_zscore(volume, 20), ts_decay_linear(returns, 20))
+
+if_else(ts_rank(returns, 30) > 0.8, ts_delta(close, 5), ts_delta(close, 60))
+
+group_rank(ts_regression(close, ts_step(1), 60, 0, 1), industry)
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 20)
+
+if_else(ts_std_dev(returns, 10) > ts_std_dev(returns, 60), ts_regression(close, ts_step(1), 10, 0, 1), ts_mean(returns, 10))
+
+multiply(rank(volume), ts_delta(close, 20))
+
+group_mean(ts_zscore(returns, 10), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(volume, 20) > 0.7, ts_delta(close, 10), 0)
+
+multiply(ts_corr(close, volume, 30), ts_delta(close, 30))
+
+ts_delta(ts_decay_linear(returns, 10), 5)
+
+if_else(ts_mean(abs(returns), 20) > 0.015, ts_delta(close, 5), ts_delta(close, 30))
+
+group_mean(ts_corr(close, returns, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+reverse(multiply(ts_rank(volume, 10), ts_rank(returns, 10)))
+
+if_else(ts_corr(close, volume, 10) > 0.3, ts_delta(close, 5), ts_delta(close, 20))
+
+group_rank(ts_zscore(volume, 30), industry)
+
+multiply(ts_std_dev(returns, 20), inverse(ts_rank(volume, 20)))
+
+ts_delta(ts_mean(close, 10), 20)
+
+if_else(rank(volume) > 0.8, ts_regression(close, ts_step(1), 20, 0, 1), ts_mean(returns, 20))
+
+group_mean(ts_decay_linear(returns, 60), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_zscore(volume, 60), ts_delta(close, 60))
+
+if_else(ts_std_dev(returns, 30) > 0.025, ts_mean(returns, 5), ts_regression(close, ts_step(1), 30, 0, 1))
+
+group_zscore(ts_delta(close, 30), industry)
+
+divide(ts_delta(close, 10), ts_mean(volume, 10))
+
+if_else(ts_rank(volume, 60) > 0.9, ts_delta(close, 5), ts_delta(close, 60))
+
+multiply(ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 60), ts_delta(close, 10))
+
+group_rank(ts_mean(returns, 20), bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_mean(abs(returns), 60) > ts_mean(abs(returns), 20), ts_delta(close, 10), ts_delta(close, 60))
+
+multiply(rank(volume), ts_mean(returns, 20))
+
+group_mean(ts_std_dev(returns, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+reverse(ts_rank(volume, 30)) * ts_regression(close, ts_step(1), 30, 0, 1)
+
+if_else(ts_corr(close, volume, 30) > 0, ts_delta(close, 20), ts_delta(close, 5))
+
+group_zscore(ts_corr(close, volume, 20), sector)
+
+ts_delta(close, 20) / ts_mean(volume, 20)
+
+if_else(bucket(rank(volume), range="0,3,0.4") == 0, ts_regression(close, ts_step(1), 30, 0, 1), ts_mean(returns, 5))
+
+multiply(ts_std_dev(returns, 30), ts_zscore(volume, 30))
+
+group_rank(ts_delta(close, 60), industry)
+
+if_else(ts_rank(abs(returns), 10) > 0.8, ts_delta(close, 5), ts_regression(close, ts_step(1), 60, 0, 1))
+
+multiply(group_mean(returns, 1, industry), ts_zscore(volume, 20))
+
+divide(ts_delta(close, 30), ts_std_dev(volume, 30))
+
+if_else(rank(volume) > 0.7, ts_decay_linear(returns, 20), ts_delta(close, 60))
+
+group_mean(ts_corr(close, returns, 60), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_rank(volume, 20), ts_rank(ts_mean(returns, 20), 20))
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.6, ts_delta(close, 10), ts_mean(returns, 30))
+
+group_zscore(ts_regression(close, ts_step(1), 20, 0, 1), sector)
+
+ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 30)
+
+if_else(volume > ts_mean(volume, 30), ts_delta(close, 5), ts_delta(close, 30))
+
+multiply(rank(volume), ts_corr(close, returns, 20))
+
+group_mean(ts_mean(returns, 10), 1, bucket(rank(volume), range="0,3,0.4"))
+
+reverse(ts_rank(volume, 60)) * ts_decay_linear(returns, 10)
+
+if_else(ts_std_dev(returns, 10) > 0.01, ts_regression(close, ts_step(1), 10, 0, 1), ts_delta(close, 60))
+
+group_rank(ts_std_dev(returns, 60), industry)
+
+multiply(ts_zscore(volume, 10), ts_zscore(returns, 10))
+
+divide(ts_delta(close, 60), ts_mean(volume, 60))
+
+if_else(bucket(rank(volume), range="0,3,0.4") == 0, ts_delta(close, 10), -ts_delta(close, 10))
+
+multiply(rank(volume), ts_delta(ts_mean(returns, 20), 5))
+
+group_mean(ts_corr(close, volume, 10), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(volume, 30) > 0.5, ts_mean(returns, 20), ts_delta(close, 10))
+
+ts_decay_linear(ts_zscore(returns, 20), 20)
+
+multiply(ts_std_dev(returns, 20), inverse(ts_mean(volume, 20)))
+
+group_zscore(ts_delta(close, 20), sector)
+
+if_else(ts_corr(close, volume, 20) < 0, ts_delta(close, 5), ts_delta(close, 30))
+
+group_rank(ts_mean(abs(returns), 10), industry)
+
+multiply(rank(returns), rank(volume))
+
+divide(ts_regression(close, ts_step(1), 30, 0, 1), ts_std_dev(returns, 30))
+
+if_else(rank(volume) > 0.6, ts_rank(returns, 10), ts_rank(returns, 60))
+
+group_mean(ts_zscore(volume, 30), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_delta(close, 20), ts_delta(volume, 20))
+
+if_else(ts_mean(volume, 10) > volume, ts_delta(close, 60), ts_delta(close, 10))
+
+group_rank(ts_regression(close, ts_step(1), 60, 0, 1), sector)
+
+reverse(multiply(ts_rank(volume, 20), ts_rank(close, 20)))
+
+if_else(ts_std_dev(returns, 60) > 0.02, ts_mean(returns, 5), ts_mean(returns, 60))
+
+group_zscore(ts_corr(close, returns, 30), industry)
+
+multiply(ts_mean(returns, 20), ts_zscore(volume, 60))
+
+divide(ts_delta(close, 5), ts_zscore(volume, 5))
+
+if_else(bucket(rank(volume), range="0,3,0.4") == 0, ts_corr(close, returns, 20), ts_corr(close, volume, 20))
+
+group_mean(ts_std_dev(returns, 10), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(rank(ts_mean(returns, 30)), rank(volume))
+
+if_else(ts_rank(ts_std_dev(returns, 20), 30) > 0.7, ts_delta(close, 10), ts_regression(close, ts_step(1), 20, 0, 1))
+
+group_rank(ts_delta(close, 10), sector)
+
+ts_corr(ts_zscore(close, 20), ts_zscore(volume, 20), 30)
+
+if_else(rank(volume) > 0.5, ts_decay_linear(returns, 20), -ts_decay_linear(returns, 20))
+
+multiply(ts_std_dev(returns, 10), ts_std_dev(volume, 10))
+
+group_mean(ts_delta(close, 60), 1, bucket(rank(volume), range="0,3,0.4"))
+
+reverse(ts_rank(volume, 10)) * ts_mean(returns, 60)
+
+divide(ts_regression(close, ts_step(1), 20, 0, 1), ts_mean(volume, 20))
+
+if_else(ts_corr(volume, returns, 10) > 0, ts_delta(close, 20), ts_delta(close, 5))
+
+group_zscore(ts_mean(returns, 10), industry)
+
+multiply(rank(volume), ts_delta(ts_decay_linear(returns, 10), 5))
+
+if_else(ts_rank(abs(returns), 20) > 0.9, ts_delta(close, 5), ts_delta(close, 30))
+
+group_rank(ts_corr(close, volume, 30), sector)
+
+multiply(ts_mean(returns, 30), inverse(ts_std_dev(returns, 30)))
+
+divide(ts_delta(close, 30), ts_delta(volume, 30))
+
+if_else(bucket(rank(volume), range="0,3,0.4") == 0, ts_delta(close, 30), ts_mean(returns, 30))
+
+group_mean(ts_zscore(returns, 60), 1, bucket(rank(volume), range="0,3,0.4"))
+
+reverse(ts_rank(volume, 20)) * ts_regression(close, ts_step(1), 60, 0, 1)
+
+if_else(ts_mean(volume, 20) > ts_mean(volume, 60), ts_delta(close, 10), ts_delta(close, 60))
+
+group_rank(ts_mean(abs(returns), 60), industry)
+
+multiply(ts_corr(close, volume, 10), ts_delta(close, 30))
+
+divide(ts_mean(returns, 20), ts_std_dev(volume, 20))
+
+if_else(rank(volume) > 0.8, ts_zscore(returns, 10), ts_zscore(returns, 60))
+
+group_mean(ts_corr(close, returns, 10), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_zscore(volume, 20), ts_zscore(close, 20))
+
+if_else(ts_std_dev(returns, 30) > ts_mean(abs(returns), 30), ts_delta(close, 5), ts_regression(close, ts_step(1), 30, 0, 1))

generated_alpha/2025-12-15/104435.txt

@@ -0,0 +1,209 @@
+ts_delta(close, 10) * rank(volume)  
+
+ts_regression(close, ts_step(1), 20, 0, 1) / ts_std_dev(returns, 20)  
+
+ts_delta(ts_delta(close, 5), 10) + ts_mean(returns, 30)  
+
+ts_mean(returns, 20) * group_mean(1, bucket(rank(volume), range="0,3,0.4"))  
+
+ts_decay_linear(returns, 30) / if_else(ts_rank(ts_std_dev(returns, 60), 120) > 0.6, 1, 0.5)  
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 60) + ts_zscore(volume, 20)  
+
+ts_delta(close, 30) * if_else(rank(volume) > 0.7, 1.2, 0.8)  
+
+group_mean(ts_delta(close, 10), 1, bucket(rank(volume), range="0,5,0.2")) - ts_std_dev(returns, 20)  
+
+ts_regression(close, ts_step(1), 60, 0, 1) * ts_mean(returns, 60)  
+
+if_else(ts_rank(volume, 20) > 0.8, ts_delta(close, 5), ts_delta(close, 60))  
+
+ts_delta(close, 20) + multiply(ts_corr(close, ts_delay(close, 1), 20), ts_std_dev(returns, 60))  
+
+group_mean(returns, 1, bucket(rank(volume), range="0,3,0.4")) * ts_delta(close, 5)  
+
+ts_delta(close, 5) / (ts_std_dev(returns, 10) + 0.0001)  
+
+ts_mean(returns, 30) * if_else(ts_zscore(volume, 20) > 1.5, 1.1, 0.9)  
+
+ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 30) * ts_regression(close, ts_step(1), 30, 0, 1)  
+
+group_rank(returns, industry) * ts_delta(close, 20)  
+
+ts_regression(close, ts_step(1), 90, 0, 1) - ts_delta(close, 60)  
+
+if_else(ts_std_dev(returns, 20) > 0.03, ts_delta(close, 5), ts_delta(close, 30))  
+
+ts_decay_linear(returns, 20) + group_mean(ts_delta(close, 5), 1, bucket(rank(volume), range="0,4,0.25"))  
+
+ts_delta(close, 60) * ts_rank(volume, 30)  
+
+ts_zscore(close, 30) * group_mean(ts_delta(close, 10), 1, bucket(rank(volume), range="0,5,0.2"))  
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 60) + ts_mean(returns, 20)  
+
+ts_delta(close, 10) * if_else(ts_rank(ts_std_dev(returns, 60), 120) > 0.5, 1.0, 0.5)  
+
+group_mean(ts_delta(close, 5), 1, bucket(rank(volume), range="0,3,0.4")) / ts_std_dev(returns, 20)  
+
+ts_regression(close, ts_step(1), 20, 0, 1) + ts_mean(returns, 60)  
+
+ts_delta(close, 30) * multiply(if_else(rank(volume) > 0.6, 1.0, 0.7), ts_rank(close, 20))  
+
+ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 30) * ts_delta(close, 5)  
+
+if_else(ts_rank(volatility, 30) > 0.7, ts_delta(close, 5), ts_delta(close, 60))  
+
+ts_delta(close, 20) + group_mean(returns, 1, bucket(rank(volume), range="0,3,0.4"))  
+
+ts_regression(close, ts_step(1), 60, 0, 1) / ts_delta(close, 10)  
+
+ts_decay_linear(returns, 90) * if_else(ts_rank(ts_std_dev(returns, 60), 120) > 0.6, 1.2, 0.8)  
+
+group_mean(ts_delta(close, 10), 1, bucket(rank(volume), range="0,5,0.2")) - ts_variance(returns, 30)  
+
+ts_delta(close, 60) * group_rank(returns, industry)  
+
+ts_zscore(ts_delta(close, 20), 60) * rank(volume)  
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 20) / ts_mean(returns, 10)  
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.5, ts_delta(close, 5), ts_mean(close, 30))  
+
+ts_mean(returns, 10) * group_mean(ts_delta(close, 20), 1, bucket(rank(volume), range="0,3,0.4"))  
+
+ts_regression(close, ts_step(1), 30, 0, 1) + ts_decay_linear(returns, 60)  
+
+ts_delta(close, 10) * multiply(1 + ts_rank(volume, 20), ts_rank(close, 60))  
+
+group_mean(close, 1, bucket(rank(volume), range="0,4,0.25")) / ts_std_dev(returns, 30)  
+
+ts_delta(close, 30) - if_else(ts_rank(volume, 20) > 0.7, 0.01, -0.01)  
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 120) * ts_regression(close, ts_step(1), 20, 0, 1)  
+
+ts_mean(returns, 60) * if_else(ts_rank(ts_std_dev(returns, 90), 180) > 0.7, 1.1, 0.9)  
+
+ts_delta(close, 5) - group_mean(ts_delta(close, 5), 1, bucket(rank(volume), range="0,5,0.2"))  
+
+ts_regression(close, ts_step(1), 90, 0, 1) * ts_zscore(volume, 30)  
+
+if_else(ts_rank(vocal, 30) > 0.6, ts_delta(close, 10), ts_delta(close, 90))  
+
+group_mean(returns, 1, bucket(rank(volume), range="0,3,0.4")) + ts_variance(returns, 60)  
+
+ts_decay_linear(returns, 20) * if_else(ts_rank(volume, 60) > 0.7, 1.0, 0.8)  
+
+ts_delta(close, 20) * group_mean(ts_regression(close, ts_step(1), 10, 0, 1), 1, bucket(rank(volume), range="0,4,0.25"))  
+
+ts_corr(ts_delta(close, 10), ts_delta(volume, 5), 30) * ts_mean(returns, 30)  
+
+ts_regression(close, ts_step(1), 60, 0, 1) - ts_delta(close, 30)  
+
+if_else(ts_rank(ts_std_dev(returns, 60), 120) > 0.7, ts_delta(close, 5), ts_delta(close, 60))  
+
+ts_delta(close, 30) + group_mean(returns, 1, bucket(rank(volume), range="0,3,0.4"))  
+
+ts_zscore(close, 60) * group_mean(ts_delta(close, 5), 1, bucket(rank(volume), range="0,5,0.2"))  
+
+ts_corr(ts_delta(close, 5), ts_delta(close, 5), 30) / ts_mean(returns, 20)  
+
+ts_delta(close, 10) * if_else(ts_rank(volume, 30) > 0.8, 1.2, 0.9)  
+
+ts_mean(returns, 60) + group_mean(close, 1, bucket(rank(volume), range="0,4,0.25"))  
+
+ts_regression(close, ts_step(1), 90, 0, 1) + ts_decay_linear(returns, 90)  
+
+ts_delta(close, 60) * multiply(if_else(rank(volume) > 0.5, 1.1, 0.95), ts_rank(close, 30))  
+
+group_mean(close, 1, bucket(rank(volume), range="0,5,0.2")) / ts_std_dev(returns, 20)  
+
+ts_delta(close, 30) - if_else(ts_rank(volume, 90) > 0.6, 0.02, -0.02)  
+
+ts_corr(ts_delta(close, 20), ts_delta(volume, 10), 60) * ts_regression(close, ts_step(1), 30, 0, 1)  
+
+ts_mean(returns, 20) * if_else(ts_rank(ts_std_dev(returns, 60), 120) > 0.5, 1.0, 0.7)  
+
+ts_delta(close, 5) + group_mean(ts_delta(close, 20), 1, bucket(rank(volume), range="0,3,0.4"))  
+
+ts_regression(close, ts_step(1), 30, 0, 1) * ts_zscore(returns, 60)  
+
+if_else(ts_rank(volume, 20) > 0.9, ts_delta(close, 5), ts_delta(close, 120))  
+
+ts_delta(close, 10) * group_mean(returns, 1, bucket(rank(volume), range="0,4,0.25"))  
+
+ts_zscore(ts_delta(close, 30), 90) * rank(volume)  
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 10), 60) / ts_mean(returns, 30)  
+
+ts_delta(close, 5) * if_else(ts_rank(volume, 60) > 0.7, 1.15, 0.9)  
+
+group_mean(close, 1, bucket(rank(volume), range="0,5,0.2")) * ts_regression(close, ts_step(1), 20, 0, 1)  
+
+ts_regression(close, ts_step(1), 20, 0, 1) * ts_mean(returns, 60)  
+
+if_else(ts_rank(ts_std_dev(returns, 30), 90) > 0.8, ts_delta(close, 5), ts_delta(close, 90))  
+
+ts_delta(close, 20) + group_mean(returns, 1, bucket(rank(volume), range="0,3,0.4"))  
+
+ts_zscore(close, 20) * group_mean(ts_delta(close, 5), 1, bucket(rank(volume), range="0,4,0.25"))  
+
+ts_corr(ts_delta(close, 10), ts_delta(volume, 5), 90) * if_else(ts_rank(volume, 30) > 0.6, 1.0, 0.8)  
+
+ts_mean(returns, 30) * if_else(ts_rank(ts_std_dev(returns, 60), 120) > 0.7, 1.2, 1.0)  
+
+ts_delta(close, 90) * rank(volume)  
+
+ts_regression(close, ts_step(1), 60, 0, 1) / ts_std_dev(returns, 10)  
+
+if_else(ts_rank(volume, 60) > 0.8, ts_delta(close, 5), ts_delta(close, 60))  
+
+ts_delta(close, 30) * group_mean(returns, 1, bucket(rank(volume), range="0,3,0.4"))  
+
+ts_zscore(close, 60) * group_mean(ts_delta(close, 10), 1, bucket(rank(volume), range="0,5,0.2"))  
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 30) * ts_mean(returns, 20)  
+
+ts_regression(close, ts_step(1), 90, 0, 1) * ts_decay_linear(returns, 120)  
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.6, ts_delta(close, 10), ts_delta(close, 120))  
+
+ts_delta(close, 20) + group_mean(ts_delta(close, 5), 1, bucket(rank(volume), range="0,4,0.25"))  
+
+ts_regression(close, ts_step(1), 30, 0, 1) + ts_zscore(returns, 90)  
+
+ts_delta(close, 60) * multiply(if_else(rank(volume) > 0.6, 1.2, 1.0), ts_rank(close, 30))  
+
+group_mean(close, 1, bucket(rank(volume), range="0,3,0.4")) / ts_std_dev(returns, 20)  
+
+ts_decay_linear(returns, 90) * if_else(ts_rank(volume, 90) > 0.7, 1.1, 0.9)  
+
+ts_delta(close, 120) + group_mean(returns, 1, bucket(rank(volume), range="0,5,0.2"))  
+
+ts_zscore(ts_delta(close, 20), 60) * rank(volume)  
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 10), 30) / ts_mean(returns, 60)  
+
+ts_delta(close, 5) * if_else(ts_rank(volume, 30) > 0.7, 1.05, 0.95)  
+
+group_mean(returns, 1, bucket(rank(volume), range="0,4,0.25")) * ts_regression(close, ts_step(1), 60, 0, 1)  
+
+ts_regression(close, ts_step(1), 20, 0, 1) * ts_mean(returns, 90)  
+
+if_else(ts_rank(ts_std_dev(returns, 60), 180) > 0.7, ts_delta(close, 5), ts_delta(close, 90))  
+
+ts_delta(close, 5) + group_mean(close, 1, bucket(rank(volume), range="0,3,0.4"))  
+
+ts_zscore(close, 30) * group_mean(ts_delta(close, 10), 1, bucket(rank(volume), range="0,5,0.2"))  
+
+ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 20) * if_else(ts_rank(volume, 90) > 0.6, 1.0, 0.7)  
+
+ts_mean(returns, 90) * if_else(ts_rank(ts_std_dev(returns, 60), 120) > 0.6, 1.1, 0.8)  
+
+ts_delta(close, 30) * rank(volume)  
+
+ts_regression(close, ts_step(1), 120, 0, 1) / ts_std_dev(returns, 30)  
+
+if_else(ts_rank(volume, 20) > 0.8, ts_delta(close, 10), ts_delta(close, 60))  
+
+ts_delta(close, 5) * group_mean(returns, 1, bucket(rank(volume), range="0,5,0.

generated_alpha/2025-12-15/104756.txt

@@ -0,0 +1,347 @@
+
+
+group_mean(ts_delta(close, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+group_mean(ts_regression(close, ts_step(1), 30, 0, 1), 1, bucket(rank(volume), range="0,2,0.5"))
+
+group_mean(ts_mean(returns, 20), 1, bucket(rank(volume), range="0,4,0.3"))
+
+ts_delta(close, 5) / ts_std_dev(returns, 20)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+ts_delta(close, if_else(ts_std_dev(returns, 20) > 0.01, 5, 60))
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, ts_delta(close, 10), ts_delta(close, 30))
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 20), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+group_mean(ts_corr(group_mean(returns, 1, industry), group_mean(returns, 1, sector), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(ts_zscore(volume, 20), ts_delta(close, 30))
+
+group_mean(ts_delta(close, 10), 1, bucket(rank(returns), range="0,3,0.5"))
+
+ts_std_dev(returns, 20) * ts_regression(close, ts_step(1), 30, 0, 1)
+
+if_else(ts_rank(volume, 30) > 0.8, mult
+
+iply(ts_delta(close, 5), volume), ts_delta(close, 60))
+
+group_mean(ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_decay_linear(ts_delta(close, 1), 30)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 120) > 0.6, ts_delta(close, 5), ts_delta(close, 120))
+
+group_mean(ts_regression(close, ts_step(1), 20, 0, 1), 1, bucket(rank(returns), range="0,4,0.2"))
+
+sign(multiply(ts_delta(close, 20), group_rank(volume, industry)))
+
+ts_delta(close, 30) / if_else(ts_rank(ts_mean(returns, 60), 120) > 0.5, 1.1, 0.9)
+
+group_mean(ts_delta(close, 10), 1, bucket(rank(volume), range="0,5,0.2"))
+
+winsorize(group_rank(returns, industry))
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 30)
+
+group_mean(ts_zscore(volume, 20), 1, bucket(rank(returns), range="0,3,0.4"))
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 60))
+
+ts_delta(close, 10) * ts_regression(ts_delta(close, 1), ts_step(1), 30, 0, 1)
+
+group_mean(multiply(ts_zscore(volume, 20), ts_delta(close, 10)), 1, bucket(rank(volume), range="0,4,0.3"))
+
+ts_decay_linear(returns, 60)
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, group_mean(returns, 1, industry), 0)
+
+group_mean(ts_corr(returns, group_mean(returns, 1, industry), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+sign(ts_delta(close, 20)) * group_mean(ts_delta(close, 10), 1, bucket(rank(volume), range="0,3,0.5"))
+
+ts_delta(close, 5) / ts_mean(returns, 30)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 120) > 0.8, ts_delta(close, 5), ts_delta(close, 60))
+
+group_mean(ts_regression(close, ts_step(1), 60, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(group_mean(ts_delta(close, 10), 1, industry))
+
+ts_zscore(multiply(ts_delta(close, 5), volume), 20)
+
+group_mean(ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+ts_delta(close, 30) * group_rank(returns, industry)
+
+group_mean(ts_delta(close, 10), 1, bucket(rank(returns), range="0,4,0.2"))
+
+norma
+
+lize(ts_delta(close, 20))
+
+ts_corr(volume, returns, 30)
+
+group_mean(if_else(rank(volume) > 0.7, close, 0), 1, industry)
+
+ts_delta(close, 5) / ts_std_dev(returns, 60)
+
+if_else(ts_rank(ts_mean(returns, 10), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 60))
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 30), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_decay_linear(ts_delta(close, 1), 5)
+
+winsorize(group_mean(rank(returns), 1, industry))
+
+group_mean(ts_corr(group_mean(returns, 1, sector), group_mean(returns, 1, market), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+sign(ts_delta(close, 10)) * if_else(ts_rank(volume, 30) > 0.8, 1, 0)
+
+ts_delta(close, 20) / if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, 1.1, 0.9)
+
+group_mean(ts_zscore(volume, 10), 1, bucket(rank(returns), range="0,3,0.5"))
+
+ts_regression(close, ts_step(1), 20, 0, 1) * group_mean(volume, 1, industry)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 120) > 0.7, ts_delta(close, 5), ts_delta(close, 60))
+
+group_mean(ts_corr(returns, volume, 30), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 10) * reversible_condition
+
+group_mean(reverse(ts_delta(close, 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 5) / ts_mean(returns, 20)
+
+if_else(ts_rank(ts_mean(returns, 30), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 120))
+
+group_mean(multiply(ts_zscore(volume, 20), ts_delta(close, 5)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+sign(group_mean(ts_delta(close, 10), 1, industry))
+
+ts_regression(ts_delta(close, 1), ts_step(1), 30, 0, 1)
+
+group_mean(ts_corr(returns, group_mean(returns, 1, sector), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 20) * if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, 1.1, 0.9)
+
+group_mean(ts_regression(close, ts_step(1), 60, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(group_mean(rank(returns), 1, industry))
+
+ts_delta(close, 10) / ts_std_dev(returns, 20)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+sign(ts_delta(close, 30)) * group_mean(ts_delta(close, 10), 1, industry)
+
+ts_decay_linear(returns, 20)
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 20), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+group_mean(ts_corr(group_mean(returns, 1, industry), group_mean(returns, 1, market), 30), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 5) / if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, 1.05, 0.95)
+
+group_mean(ts_regression(close, ts_step(1), 30, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(group_mean(ts_delta(close, 20), 1, industry))
+
+ts_zscore(multiply(ts_delta(close, 10), volume), 20)
+
+group_mean(ts_corr(returns, volume, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+ts_delta(close, 10) * group_rank(returns, industry)
+
+group_mean(ts_mean(returns, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_regression(close, ts_step(1), 20, 0, 1)
+
+group_mean(ts_corr(group_mean(returns, 1, sector), group_mean(returns, 1, market), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+sign(group_mean(ts_delta(close, 10), 1, industry))
+
+ts_decay_linear(ts_delta(close, 1), 30)
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 30), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 5) / ts_mean(returns, 60)
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 120))
+
+group_mean(multiply(ts_zscore(volume, 10), ts_delta(close, 5)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+norma
+
+lize(group_mean(ts_delta(close, 10), 1, industry))
+
+ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 30)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(ts_regression(close, ts_step(1), 60, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(sign(ts_delta(close, 10)))
+
+ts_zscore(multiply(ts_delta(close, 5), volume), 20)
+
+group_mean(ts_corr(returns, group_mean(returns, 1, sector), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(ts_len
+
+g(volume), 30) > 0.8, ts_delta(close, 5), ts_delta(close, 30))
+
+sign(ts_delta(close, 20)) * group_mean(ts_delta(close, 10), 1, industry)
+
+ts_decay_linear(returns, 30)
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 20), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 10) / ts_std_dev(returns, 20)
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 60))
+
+group_mean(multiply(ts_zscore(volume, 20), ts_delta(close, 5)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+norma
+
+lize(group_mean(rank(returns), 1, industry))
+
+ts_corr(ts_delta(close, 5), ts_delta(volume, 5), 20)
+
+if_else(ts_rank(volume, 30) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(ts_regression(close, ts_step(1), 30, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(group_mean(ts_mean(returns, 20), 1, industry))
+
+ts_zscore(group_mean(volume, 1, sector), 20)
+
+group_mean(ts_corr(returns, volume, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+sign(group_mean(ts_delta(close, 10), 1, industry))
+
+ts_delta(close, 5) / ts_mean(returns, 30)
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 10), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_regression(close, ts_step(1), 20, 0, 1)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 120) > 0.6, ts_delta(close, 5), ts_delta(close, 120))
+
+group_mean(ts_delta(close, 20), 1, bucket(rank(returns), range="0,3,0.4"))
+
+winsorize(group_mean(ts_delta(close, 10), 1, industry))
+
+ts_zscore(
+
+multiply(ts_delta(close, 10), ts_delta(volume, 10)), 20)
+
+group_mean(ts_corr(group_mean(returns, 1, industry), group_mean(returns, 1, market), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 60))
+
+sign(group_mean(rank(returns), 1, industry))
+
+ts_decay_linear(returns, 20)
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 30), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 10) / ts_std_dev(returns, 30)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(ts_regression(close, ts_step(1), 60, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(group_mean(rank(returns), 1, industry))
+
+ts_zscore(multiply(ts_delta(close, 5), volume), 20)
+
+group_mean(ts_corr(returns, group_mean(returns, 1, sector), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 120))
+
+group_mean(ts_delta(close, 5), 1, bucket(rank(volume), range="0,3,0.4"))
+
+sign(group_mean(ts_delta(close, 10), 1, industry))
+
+ts_decay_linear(ts_delta(close, 1), 30)
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 20), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 20) / ts_mean(returns, 20)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 30))
+
+group_mean(ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(sign(ts_delta(close, 10)))
+
+ts_zscore(group_mean(volume, 1, industry), 20)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(ts_regression(close, ts_step(1), 30, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(group_mean(ts_mean(returns, 20), 1, industry))
+
+ts_zscore(group_mean(multiply(ts_delta(close, 5), volume), 1, sector), 20)
+
+group_mean(ts_corr(returns, group_mean(returns, 1, industry), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 120))
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 10), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 10) / ts_std_dev(returns, 20)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(ts_regression(close, ts_step(1), 20, 0, 1), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(group_mean(rank(returns), 1, industry))
+
+ts_zscore(multiply(ts_delta(close, 5), ts_delta(volume, 5)), 20)
+
+group_mean(ts_corr(returns, volume, 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+multiply(if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, 1.1, 0.9), ts_delta(close, 20))
+
+group_mean(ts_corr(group_mean(returns, 1, sector), group_mean(returns, 1, market), 20), 1, bucket(rank(returns), range="0,3,0.4"))
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 60))
+
+sign(group_mean(ts_delta(close, 10), 1, industry))
+
+ts_decay_linear(returns, 20)
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 30), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 5) / ts_mean(returns, 60)
+
+if_else(ts_rank(ts_std_dev(returns, 20), 60) > 0.7, ts_delta(close, 5), ts_delta(close, 20))
+
+group_mean(ts_corr(ts_delta(close, 10), ts_delta(volume, 10), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+winsorize(sign(group_mean(ts_delta(close, 10), 1, industry)))
+
+ts_zscore(group_mean(volume, 1, sector), 20)
+
+group_mean(ts_corr(returns, group_mean(returns, 1, industry), 20), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_rank(ts_mean(returns, 20), 60) > 0.5, ts_delta(close, 5), ts_delta(close, 60))
+
+group_mean(reverse(ts_rank(volume/ts_mean(volume, 20), 10)), 1, bucket(rank(volume), range="0,3,0.4"))
+
+ts_delta(close, 20) / ts_std_dev(returns, 20)

generated_alpha/2025-12-16/110013.txt

@@ -0,0 +1,39 @@
+group_rank(if_else(ts_zscore(volume, 20) > 0, ts_delta(close, 10), reverse(ts_delta(close, 10))), sector)
+
+if_else(bucket(rank(close * volume), range="0,3,0.4") == 0, ts_regression(close, ts_step(1), 30, 0, 1), group_mean(ts_delta(close, 20), 1, sector))
+
+multiply(group_zscore(volume, sector), ts_std_dev(returns, 60))
+
+multiply(ts_rank(ts_std_dev(returns, 20), 120), if_else(ts_zscore(volume, 5) > 1, ts_delta(close, 5), ts_delta(close, 20)))
+
+ts_delta(group_rank(ts_sum(volume, 5), sector), 10)
+
+group_mean(multiply(ts_corr(returns, ts_sum(volume, 10), 30), if_else(ts_zscore(close, 20) > 0, 1, -1)), 1, sector)
+
+if_else(ts_rank(volume / ts_mean(volume, 20), 10) > 0.7, reverse(ts_delta(close, 5)), ts_delta(close, 20))
+
+multiply(group_zscore(ts_delta(close, 5), sector), reverse(ts_rank(ts_std_dev(returns, 60), 120)))
+
+ts_delta(group_mean(if_else(ts_mean(returns, 20) > 0, ts_zscore(volume, 20), reverse(ts_zscore(volume, 20))), 1, industry), 30)
+
+multiply(ts_corr(ts_delta(close, 5), ts_delta(close, 20), 60), group_rank(ts_sum(volume, 5), sector))
+
+group_scale(ts_av_diff(close, 30), sector)
+
+if_else(ts_std_dev(returns, 20) > ts_std_dev(returns, 60), ts_delta(close, 10), group_mean(ts_delta(close, 30), 1, industry))
+
+multiply(ts_rank(ts_delta(volume, 5), 20), group_zscore(ts_regression(close, ts_step(1), 30, 0, 1), sector))
+
+ts_delta(group_mean(if_else(rank(volume) > 0.8, ts_delta(close, 10), ts_delta(close, 5)), 1, sector), 20)
+
+subtract(group_rank(ts_sum(volume, 10), sector), ts_rank(ts_std_dev(returns, 30), 120))
+
+multiply(ts_corr(group_mean(returns, 1, sector), ts_delta(close, 10), 30), ts_delta(ts_sum(volume, 5), 20))
+
+if_else(bucket(rank(ts_std_dev(returns, 60)), range="0,3,0.4") == 2, ts_delta(close, 5), group_mean(ts_delta(close, 20), 1, industry))
+
+group_zscore(multiply(ts_delta(close, 5), if_else(ts_zscore(volume, 10) > 0, 1, -1)), sector)
+
+ts_delta(group_rank(ts_mean(abs(returns), 20), industry), 10)
+
+multiply(reverse(ts_rank(volume / ts_mean(volume, 30), 20)), group_mean(ts_delta(close, 10), 1, sector))

generated_alpha/2025-12-16/110049.txt

@@ -0,0 +1,39 @@
+multiply(ts_delta(close,20), ts_zscore(volume,30))
+
+group_mean(ts_delta(close,10), 1, bucket(rank(volume), range="0,3,0.4"))
+
+if_else(ts_std_dev(returns,20) > ts_quantile(ts_std_dev(returns,20),120,"gaussian"), ts_delta(close,5), ts_delta(close,30))
+
+ts_corr(ts_delta(close,5), ts_delta(volume,5), 60)
+
+multiply(ts_regression(close, ts_step(1), 30, 0, 1), group_rank(ts_delta(close,20), sector))
+
+ts_av_diff(ts_zscore(volume,20), 10)
+
+group_neutralize(ts_delta(close,60), industry)
+
+multiply(ts_rank(ts_delta(close,10), 60), ts_zscore(volume, 20))
+
+if_else(ts_rank(ts_std_dev(returns,60), 120) > 0.7, ts_delta(close,10), ts_delta(close,60))
+
+group_zscore(ts_delta(close,30), sector)
+
+multiply(ts_delta(close,20), ts_delta(volume,20))
+
+ts_decay_linear(ts_delta(close,5), 30)
+
+group_scale(ts_zscore(volume,60), industry)
+
+multiply(ts_corr(group_mean(returns,1,sector), group_mean(returns,1,industry), 30), ts_delta(close,20))
+
+ts_regression(ts_delta(close,1), ts_step(1), 30, 0, 1)
+
+if_else(bucket(rank(volume), range="0,3,0.4") == 0, ts_delta(close,10), ts_delta(close,30))
+
+ts_mean(ts_delta(close,1), 20)
+
+group_mean(ts_zscore(volume,30), 1, sector)
+
+multiply(ts_rank(ts_delta(close,20), 60), ts_std_dev(returns,20))
+
+ts_scale(ts_delta(close,60), 120)

generated_alpha/2025-12-16/110056.txt

@@ -0,0 +1,21 @@
+
+multiply(ts_corr(group_mean(returns,1,sector), group_mean(returns,1,industry),30), ts_delta(close,20))
+divide(ts_zscore(volume,20), add(ts_std_dev(returns,30), 0.01))
+multiply(ts_delta(close,10), if_else(rank(volume) > 0.8, 1, -1))
+ts_corr(rank(ts_delta(close,5)), rank(ts_zscore(volume,10)), 20)
+multiply(ts_delta/ts_std_dev(ts_delta(close,20),20), -1)
+multiply(group_rank(close,industry), rank(volume))
+ts/ts_mean(ts_mean(volume,5),20)
+multiply(-1, ts_rank(rank(close),10))
+divide(subtract(mean(close,10), mean(close,50)), ts_std_dev(returns,100))
+multiply(hump(mean(close,10),0.05), hump(mean(close,50),0.05))
+multiply(rank(volume), -1 * ts_delta(close,30))
+multiply(-1, rank(ts_std_dev(returns,60)))
+divide(add(close,vwap),2)
+multiply(subtract(close,ts_delay(close,30)), -1 * ts_std_dev(returns,60))
+multiply(hump(mean(close,30),0.1), rank(volume))
+multiply(if_else(ts_std_dev(returns,60) > 0.02, -1 * ts_delta(mean(close,5),5), -1 * ts_delta(mean(close,20),20)), -1 * rank(volume))
+multiply(if_else(close > ts_delay(close,30), 1, -1), rank(volume))
+multiply(ts_corr(close, volume,20), -1 * ts_delta(close,5))
+multiply(hump(mean(close,15),0.05), hump(mean(close,60),0.1))
+multiply(if_else(ts_std_dev(returns,100) > 0.015, close/min(close,100), 0), hump(volume,0.1))

generated_alpha/2025-12-16/110120.txt

@@ -0,0 +1,39 @@
+close / ts_delay(close, 20) - 1 + ts_zscore(volume, 30) * (ts_delta(close, 5) / abs(ts_delta(close, 5))) * if_else(ts_rank(ts_std_dev(returns, 20), 200) > 0.7, 1, -1)
+
+ts_delta(close, 5) / ts_delay(close, 5) - 1 + bucket(rank(ts_delta(close, 10)), range="0,5,0.2") * ts_zscore(volume, 10)
+
+ts_regression(close, ts_step(1), 30, 0, 1) * (volume / ts_mean(volume, 30)) / ts_std_dev(returns, 10)
+
+(ts_delta(close, 10) / ts_delay(close, 10) - 1) * group_mean(volume / ts_mean(volume, 60), 1, industry) + if_else(ts_rank(close, 100) < 0.3, 1, 0)
+
+ts_mean(returns, 5) - ts_mean(returns, 20) + ts_zscore(volume, 20) * if_else(ts_rank(volume, 200) > 0.8, 1, -1) * bucket(ts_corr(close, close / ts_delay(close, 30), 30), range="0,4,0.5")
+
+ts_delta(close, 20) / ts_delay(close, 20) - 1 + (ts_rank(volume, 100) / 100) * if_else(ts_rank(close, 120) < 0.5, ts_delta(close, 5), -ts_delta(close, 5))
+
+group_zscore(close / ts_delay(close, 30) - 1, sector) * ts_zscore(volume, 10) - ts_std_dev(returns, 20) * if_else(ts_rank(volume, 100) > 0.9, 1, 0)
+
+ts_corr(close, volume, 20) * (ts_delta(close, 5) / ts_delay(close, 5) - 1) + bucket(ts_zscore(returns, 60), range="0,5,0.5") * if_else(ts_rank(volume, 120) < 0.4, -1, 1)
+
+ts_mean(close, 5) / ts_mean(close, 60) - 1 + group_mean(ts_delta(close, 1) / ts_delay(close, 1) - 1, 1, industry) * ts_zscore(volume / ts_mean(volume, 30), 30)
+
+(ts_delta(close, 20) / ts_delay(close, 20) - 1) * (volume / ts_mean(volume, 10)) + if_else(ts_rank(ts_std_dev(returns, 30), 100) > 0.6, ts_mean(returns, 20), 0)
+
+ts_delta(close, 5) / ts_delay(close, 5) - 1 + ts_zscore(volume, 60) * if_else(ts_rank(ts_corr(close, close / ts_delay(close, 20), 20), 100) > 0.7, 1, -1)
+
+ts_rank(volume, 100) / 100 * (ts_delta(close, 10) / ts_delay(close, 10) - 1) - bucket(ts_std_dev(returns, 20), range="0,4,0.5") * if_else(ts_rank(close, 200) < 0.3, 1, 0)
+
+ts_regression(close, ts_step(1), 20, 0, 1) * (volume / ts_mean(volume, 20)) + group_mean(ts_zscore(returns, 30), 1, sector) * if_else(ts_rank(close, 100) > 0.7, 1, -1)
+
+(ts_delta(close, 30) / ts_delay(close, 30) - 1) + ts_zscore(volume, 10) * bucket(ts_rank(ts_std_dev(returns, 10), 200), range="0,3,0.3")
+
+close / ts_delay(close, 60) - 1 + group_mean(ts_delta(close, 5), 1, industry) * if_else(ts_rank(volume, 200) > 0.9, 1, -1)
+
+ts_corr(close / ts_delay(close, 5) - 1, volume / ts_mean(volume, 5), 20) * ts_delta(close, 1) / ts_delay(close, 1) - bucket(rank(volume), range="0,3,0.3")
+
+ts_mean(returns, 10) - ts_mean(returns, 60) + group_zscore(volume, sector) * if_else(ts_rank(close, 50) < 0.4, -1, 1)
+
+(ts_delta(close, 5) / ts_delay(close, 5) - 1) * (ts_corr(close, ts_mean(close, 20), 20) + 1) / 2 + if_else(ts_rank(volume, 150) > 0.7, ts_zscore(returns, 20), 0)
+
+ts_delta(close, 20) / ts_delay(close, 20) - 1 + bucket(ts_rank(volume, 100), range="0,4,0.25") * group_mean(ts_delta(close, 1), 1, industry)
+
+ts_rank(volume, 100) / 100 * (ts_delta(close, 20) / ts_delay(close, 20) - 1) + if_else(ts_rank(ts_std_dev(returns, 60), 200) > 0.5, group_mean(returns, 1, sector), -group_mean(returns, 1, sector))

generated_alpha/2025-12-16/110532.txt

@@ -0,0 +1,19 @@
+group_mean(multiply(ts_zscore(volume, 5), group_mean(abs(ts_delta(close, 5)), 1, industry)), 1, bucket(rank(volume), range="0,3,0.4"))
+multiply(group_mean(ts_delta(close, 10), 1, industry), ts_regression(volume, ts_step(1), 30, 0, 1))
+ts_corr(multiply(volume, returns), group_mean(multiply(volume, returns), 1, sector), 20)
+group_mean(if_else(ts_std_dev(volume, 10) < 0.7, ts_delta(close, 5), -ts_delta(close, 5)), 1, bucket(rank(volume), range="0,3,0.4"))
+multiply(ts_zscore(volume, 15), divide(group_mean(ts_delta(close, 1), 1, industry), ts_std_dev(returns, 15)))
+group_mean(ts_delta(ts_regression(close, ts_step(1), 20, 0, 1), 1, industry), 1, bucket(rank(ts_regression(close, ts_step(1), 30, 0, 1), range="0,3,0.4", 2)))
+multiply(group_mean(days_from_last_change(high_volume_event), 1, sector), ts_zscore(volume, 10))
+group_mean(ts_delay(close, 1) / close - group_mean(ts_delay(close, 1) / close, 1, industry), 1, bucket(rank(volume), range="0,3,0.4"))
+ts_corr(ts_delta(close, 1), group_mean(ts_delta(close, 1), 1, sector), 30)
+divide(group_mean(ts_delta(close, 20), 1, industry), ts_std_dev(returns, 20))
+multiply(ts_zscore(volume, 5), group_mean(ts_delta(close, 1), 1, bucket(rank(volume), range="0,3,0.4", 1)))
+group_mean(ts_delay(multiply(volume, returns), 1), 1, industry) * ts_regression(close, ts_step(1), 60, 0, 1)
+divide(group_mean(ts_delta(volume / sharesout, 1), 1, industry_leader_bucket), group_mean(volume / sharesout, 1, industry))
+multiply(group_mean(abs(returns - group_mean(returns, 1, industry)), 1, industry), ts_zscore(volume, 10))
+group_mean(ts_delta(close, 5), 1, bucket(rank(ts_regression(close, ts_step(1), 20, 0, 1), range="0,3,0.4")))
+if_else(group_mean(ts_corr(volume, group_mean(close, 1, industry), 20), 1, industry) > 0.6, ts_delta(close, 5), -ts_delta(close, 5))
+group_mean(days_from_last_change(ts_zscore(volume, 5)), 1, industry) * ts_std_dev(returns, 20)
+multiply(ts_zscore(volume, 10), group_mean(ts_delta(close, 1), 1, bucket(rank(volume), range="0,3,0.5")))
+divide(group_mean(ts_delta(close, 10), 1, industry), ts_std_dev(returns, 10))

main.py

@@ -152,22 +152,21 @@ def manual_prompt(prompt):
     print(f"手动提示词保存到: {filepath}")
 
 
-def call_ai(prompt):
+def call_ai(prompt, model):
     balance = get_user_info()
 
     folder = create_result_folder()
 
-    for model in MODELS:
-        print(f"正在调用AI...{model}")
-        result = call_siliconflow(prompt, model)
-
-        if result:
-            print(f"AI回复: {result[:200]}...")
-            save_result(result, folder)
-            used_balance = balance - get_user_info()
-            print(f'本次调用 api 使用额度 {used_balance}')
-        else:
-            print("AI调用失败")
+    print(f"正在调用AI...{model}")
+    result = call_siliconflow(prompt, model)
+
+    if result:
+        print(f"AI回复: {result[:200]}...")
+        save_result(result, folder)
+        used_balance = balance - get_user_info()
+        print(f'本次调用 api 使用额度 {used_balance}')
+    else:
+        print("AI调用失败")
 
 
 def prepare_prompt():
@@ -220,13 +219,14 @@ def prepare_prompt():
 
 
 def main():
-    prompt = prepare_prompt()
+    for model in MODELS:
+        prompt = prepare_prompt()
 
-    # # 如果需要手动在页面段模型, 使用提示词, 打开这个, 将生成的提示词存到本地
-    manual_prompt(prompt)
+        # # 如果需要手动在页面段模型, 使用提示词, 打开这个, 将生成的提示词存到本地
+        manual_prompt(prompt)
 
-    # # 如果需要使用模型, 打开这个
-    call_ai(prompt)
+        # # 如果需要使用模型, 打开这个
+        call_ai(prompt, model)
 
 
 if __name__ == "__main__":

manual_prompt/manual_prompt_20251215103723.txt

@@ -0,0 +1,898 @@
+任务指令
+你是一个WorldQuant WebSim因子工程师。你的任务是生成 100 个用于行业轮动策略的复合型Alpha因子表达式。
+核心规则
+设计维度框架
+维度1：时间序列动量（TM）
+目标：识别价格趋势的强度、速度和持续性
+可用的具体构建方法：
+1. 简单动量：ts_delta(close, d) [d=5,10,20,30,60]
+2. 趋势斜率：ts_regression(close, ts_step(1), d, 0, 1) [rettype=1获取斜率]
+3. 动量加速度：ts_delta(ts_delta(close, d1), d2) [避免嵌套ts_regression]
+4. 平滑动量：ts_mean(returns, d) [returns=ts_delta(close,1)]
+5. 动量衰减：ts_decay_linear(returns, d)
+6. 价量关系：ts_corr(ts_delta(close,5), ts_delta(volume,5), d)
+建议组合：使用不同d参数创建短期/中期/长期动量
+维度2：横截面领导力（CL）
+目标：识别行业内的龙头股和相对强度
+具体构建方法：
+1. 龙头股筛选：if_else(rank(volume) > 0.7, 龙头值, 其他值) [使用volume代替market_cap]
+2. 龙头组合：group_mean(x, 1, bucket(rank(volume), range="0,3,0.4")) [使用volume排序]
+3. 行业内离散度：ts_std_dev(group_rank(returns, industry), 20)
+4. 相对排名稳定性：ts_mean(rank(returns), d)
+维度3：市场状态适应性（MS）
+目标：根据波动率、趋势状态调整参数
+具体构建方法：
+1. 波动率调整：ts_delta(close,5) / ts_std_dev(returns,20)
+2. 状态条件选择：if_else(ts_rank(volatility,30) > 0.7, 短期动量, 长期动量)
+3. 参数动态化：if_else(ts_std_dev(returns,20) > 阈值, 5, 20) [作为d参数]
+4. 趋势状态识别：ts_rank(ts_mean(returns,20), 60) > 0.5
+基本结构：
+复合因子 = 维度A组件 [运算符] 维度B组件 [条件调整]
+=== 关键语法规则（必须遵守） ===
+1. 数据字段规范：
+   - 可使用字段：close, volume, returns
+   - ❌ 错误：market_cap, marketcap, mkt_cap [这些字段不存在]
+   - ✅ 正确：使用volume作为规模代理，close作为价格
+   - returns通常定义为：ts_delta(close, 1) 或 close/ts_delay(close,1)-1
+2. ts_regression使用规范：
+   - 避免深度嵌套ts_regression，特别是作为其他函数的参数
+   - ✅ 正确：reg_slope = ts_regression(close, ts_step(1), 30, 0, 1)
+   - ❌ 错误：ts_delta(ts_regression(close, ts_step(1), 30, 0, 1), 5)
+   - 替代方案：用ts_delta组合计算动量变化
+3. if_else使用规范：
+   - 条件必须是简单布尔表达式
+   - 避免序列比较：❌ ts_std_dev(returns,60) > ts_mean(ts_std_dev(returns,60),120)
+   - 正确使用：✅ if_else(ts_rank(ts_std_dev(returns,60), 120) > 0.7, 短期动量, 长期动量)
+4. bucket函数使用规范：
+   - bucket()返回分组ID，可用于条件判断
+   - ✅ 正确：bucket(rank(volume), range="0,3,0.4") == 0 [第一组为大成交量]
+   - ✅ 正确：group_mean(x, 1, bucket(rank(volume), range="0,3,0.4"))
+   - 注意字符串格式：range="起始值,组数,步长" 或 buckets="分割点列表"
+=== 关键语法规则结束 ===
+*=====*
+注意事项:
+1. 避免过度复杂的嵌套（建议不超过3层）
+2. 每个表达式应有明确的经济逻辑
+3. 考虑实际交易可行性（避免未来函数）
+4. 包含风险控制元素（如波动率调整）
+5. 只能使用可用的数据字段：close, volume, returns等
+*=====*
+参数逻辑：参数d（回顾期）应在[5, 10, 20, 30, 60, 120]等具有市场意义（周、月、季度、半年）的数值中合理选择并差异化。
+行业隐含：通过group_mean、group_rank等函数或假设表达式在行业指数上运行来体现"行业"逻辑。
+构建框架指导（请按此逻辑创造新因子）：
+维度融合模板（选择至少2个）：
+A. 领导力动量 = 时序动量 × 横截面调整
+   逻辑：大成交量股票的动量更强
+   结构：group_mean(ts_delta(close, d1), 1, bucket(rank(volume), range="0,3,0.4"))
+B. 状态自适应动量 = 条件选择动量
+   逻辑：高波动用短期动量，低波动用长期动量
+   结构：if_else(ts_std_dev(returns,20) > 0.02, ts_delta(close,5), ts_delta(close,20))
+C. 行业传导因子 = 领先行业动量 × 相关性强度
+   逻辑：与强势行业相关性高的行业未来表现好
+   结构：multiply(ts_corr(group_mean(returns,1,industry), group_mean(returns,1,sector), d1), ts_delta(close,d2))
+D. 情绪反转 = 过度交易信号 × 基础趋势
+   逻辑：过度交易时反转，趋势延续时跟随
+   结构：multiply(reverse(ts_rank(volume/ts_mean(volume,20), 10)), ts_delta(close,20))
+关键组件库（可自由组合）：
+1. 动量类：ts_delta(close,{d}), ts_regression(close,ts_step(1),{d},0,1)
+2. 波动类：ts_std_dev(returns,{d}), ts_mean(abs(returns),{d})
+3. 成交量类：volume/ts_mean(volume,{d}), ts_zscore(volume,{d})
+4. 横截面类：if_else(rank(volume) > 阈值, 值1, 值2), bucket(rank(volume), range="0,3,0.4")
+5. 相关性类：ts_corr({x},{y},{d})
+6. 条件逻辑：if_else({condition}, {true_value}, {false_value})
+参数池：d ∈ [5,10,20,30,60,120], 阈值 ∈ [0.5,0.7,0.8]
+*=====*
+输出格式：
+输出必须是且仅是纯文本。
+每一行是一个完整、独立、语法正确的WebSim表达式。
+严禁任何形式的解释、编号、标点包裹（如引号）、Markdown格式或额外文本。
+===================== !!! 重点(输出方式) !!! =====================
+现在，请严格遵守以上所有规则，开始生成可立即在WebSim中运行的复合因子表达式。
+**输出格式**(一行一个表达式, 每个表达式中间需要添加一个空行, 只要表达式本身, 不要解释, 不需要序号, 也不要输出多余的东西)：
+表达式
+表达式
+表达式
+...
+表达式
+=================================================================
+重申：请确保所有表达式都使用WorldQuant WebSim平台函数，不要使用pandas、numpy或其他Python库函数。输出必须是一行有效的WQ表达式。
+以下是我的账号有权限使用的操作符, 请严格按照操作符, 进行生成,组合因子：
+
+以下是我的账号有权限使用的操作符, 请严格按照操作符, 进行生成,组合因子
+
+========================= 操作符开始 =======================================注意: Operator: 后面的是操作符,
+Description: 此字段后面的是操作符对应的描述或使用说明, Description字段后面的内容是使用说明, 不是操作符
+特别注意!!!! 必须按照操作符字段Operator的使用说明生成 alphaOperator: abs(x)
+Description: Absolute value of x
+Operator: add(x, y, filter = false)
+Description: Add all inputs (at least 2 inputs required). If filter = true, filter all input NaN to 0 before adding
+Operator: densify(x)
+Description: Converts a grouping field of many buckets into lesser number of only available buckets so as to make working with grouping fields computationally efficient
+Operator: divide(x, y)
+Description: x / y
+Operator: inverse(x)
+Description: 1 / x
+Operator: log(x)
+Description: Natural logarithm. For example: Log(high/low) uses natural logarithm of high/low ratio as stock weights.
+Operator: max(x, y, ..)
+Description: Maximum value of all inputs. At least 2 inputs are required
+Operator: min(x, y ..)
+Description: Minimum value of all inputs. At least 2 inputs are required
+Operator: multiply(x ,y, ... , filter=false)
+Description: Multiply all inputs. At least 2 inputs are required. Filter sets the NaN values to 1
+Operator: power(x, y)
+Description: x ^ y
+Operator: reverse(x)
+Description: - x
+Operator: sign(x)
+Description: if input > 0, return 1; if input < 0, return -1; if input = 0, return 0; if input = NaN, return NaN;
+Operator: signed_power(x, y)
+Description: x raised to the power of y such that final result preserves sign of x
+Operator: sqrt(x)
+Description: Square root of x
+Operator: subtract(x, y, filter=false)
+Description: x-y. If filter = true, filter all input NaN to 0 before subtracting
+Operator: and(input1, input2)
+Description: Logical AND operator, returns true if both operands are true and returns false otherwise
+Operator: if_else(input1, input2, input 3)
+Description: If input1 is true then return input2 else return input3.
+Operator: input1 < input2
+Description: If input1 < input2 return true, else return false
+Operator: input1 <= input2
+Description: Returns true if input1 <= input2, return false otherwise
+Operator: input1 == input2
+Description: Returns true if both inputs are same and returns false otherwise
+Operator: input1 > input2
+Description: Logic comparison operators to compares two inputs
+Operator: input1 >= input2
+Description: Returns true if input1 >= input2, return false otherwise
+Operator: input1!= input2
+Description: Returns true if both inputs are NOT the same and returns false otherwise
+Operator: is_nan(input)
+Description: If (input == NaN) return 1 else return 0
+Operator: not(x)
+Description: Returns the logical negation of x. If x is true (1), it returns false (0), and if input is false (0), it returns true (1).
+Operator: or(input1, input2)
+Description: Logical OR operator returns true if either or both inputs are true and returns false otherwise
+Operator: days_from_last_change(x)
+Description: Amount of days since last change of x
+Operator: hump(x, hump = 0.01)
+Description: Limits amount and magnitude of changes in input (thus reducing turnover)
+Operator: kth_element(x, d, k)
+Description: Returns K-th value of input by looking through lookback days. This operator can be used to backfill missing data if k=1
+Operator: last_diff_value(x, d)
+Description: Returns last x value not equal to current x value from last d days
+Operator: ts_arg_max(x, d)
+Description: Returns the relative index of the max value in the time series for the past d days. If the current day has the max value for the past d days, it returns 0. If previous day has the max value for the past d days, it returns 1
+Operator: ts_arg_min(x, d)
+Description: Returns the relative index of the min value in the time series for the past d days; If the current day has the min value for the past d days, it returns 0; If previous day has the min value for the past d days, it returns 1.
+Operator: ts_av_diff(x, d)
+Description: Returns x - tsmean(x, d), but deals with NaNs carefully. That is NaNs are ignored during mean computation
+Operator: ts_backfill(x,lookback = d, k=1, ignore="NAN")
+Description: Backfill is the process of replacing the NAN or 0 values by a meaningful value (i.e., a first non-NaN value)
+Operator: ts_corr(x, y, d)
+Description: Returns correlation of x and y for the past d days
+Operator: ts_count_nans(x ,d)
+Description: Returns the number of NaN values in x for the past d days
+Operator: ts_covariance(y, x, d)
+Description: Returns covariance of y and x for the past d days
+Operator: ts_decay_linear(x, d, dense = false)
+Description: Returns the linear decay on x for the past d days. Dense parameter=false means operator works in sparse mode and we treat NaN as 0. In dense mode we do not.
+Operator: ts_delay(x, d)
+Description: Returns x value d days ago
+Operator: ts_delta(x, d)
+Description: Returns x - ts_delay(x, d)
+Operator: ts_mean(x, d)
+Description: Returns average value of x for the past d days.
+Operator: ts_product(x, d)
+Description: Returns product of x for the past d days
+Operator: ts_quantile(x,d, driver="gaussian" )
+Description: It calculates ts_rank and apply to its value an inverse cumulative density function from driver distribution. Possible values of driver (optional ) are "gaussian", "uniform", "cauchy" distribution where "gaussian" is the default.
+Operator: ts_rank(x, d, constant = 0)
+Description: Rank the values of x for each instrument over the past d days, then return the rank of the current value + constant. If not specified, by default, constant = 0.
+Operator: ts_regression(y, x, d, lag = 0, rettype = 0)
+Description: Returns various parameters related to regression function
+Operator: ts_scale(x, d, constant = 0)
+Description: Returns (x - ts_min(x, d)) / (ts_max(x, d) - ts_min(x, d)) + constant. This operator is similar to scale down operator but acts in time series space
+Operator: ts_std_dev(x, d)
+Description: Returns standard deviation of x for the past d days
+Operator: ts_step(1)
+Description: Returns days' counter
+Operator: ts_sum(x, d)
+Description: Sum values of x for the past d days.
+Operator: ts_zscore(x, d)
+Description: Z-score is a numerical measurement that describes a value's relationship to the mean of a group of values. Z-score is measured in terms of standard deviations from the mean: (x - tsmean(x,d)) / tsstddev(x,d). This operator may help reduce outliers and drawdown.
+Operator: normalize(x, useStd = false, limit = 0.0)
+Description: Calculates the mean value of all valid alpha values for a certain date, then subtracts that mean from each element
+Operator: quantile(x, driver = gaussian, sigma = 1.0)
+Description: Rank the raw vector, shift the ranked Alpha vector, apply distribution (gaussian, cauchy, uniform). If driver is uniform, it simply subtract each Alpha value with the mean of all Alpha values in the Alpha vector
+Operator: rank(x, rate=2)
+Description: Ranks the input among all the instruments and returns an equally distributed number between 0.0 and 1.0. For precise sort, use the rate as 0
+Operator: scale(x, scale=1, longscale=1, shortscale=1)
+Description: Scales input to booksize. We can also scale the long positions and short positions to separate scales by mentioning additional parameters to the operator
+Operator: winsorize(x, std=4)
+Description: Winsorizes x to make sure that all values in x are between the lower and upper limits, which are specified as multiple of std.
+Operator: zscore(x)
+Description: Z-score is a numerical measurement that describes a value's relationship to the mean of a group of values. Z-score is measured in terms of standard deviations from the mean
+Operator: vec_avg(x)
+Description: Taking mean of the vector field x
+Operator: vec_sum(x)
+Description: Sum of vector field x
+Operator: bucket(rank(x), range="0, 1, 0.1" or buckets = "2,5,6,7,10")
+Description: Convert float values into indexes for user-specified buckets. Bucket is useful for creating group values, which can be passed to GROUP as input
+Operator: trade_when(x, y, z)
+Description: Used in order to change Alpha values only under a specified condition and to hold Alpha values in other cases. It also allows to close Alpha positions (assign NaN values) under a specified condition
+Operator: group_backfill(x, group, d, std = 4.0)
+Description: If a certain value for a certain date and instrument is NaN, from the set of same group instruments, calculate winsorized mean of all non-NaN values over last d days
+Operator: group_mean(x, weight, group)
+Description: All elements in group equals to the mean
+Operator: group_neutralize(x, group)
+Description: Neutralizes Alpha against groups. These groups can be subindustry, industry, sector, country or a constant
+Operator: group_rank(x, group)
+Description: Each elements in a group is assigned the corresponding rank in this group
+Operator: group_scale(x, group)
+Description: Normalizes the values in a group to be between 0 and 1. (x - groupmin) / (groupmax - groupmin)
+Operator: group_zscore(x, group)
+Description: Calculates group Z-score - numerical measurement that describes a value's relationship to the mean of a group of values. Z-score is measured in terms of standard deviations from the mean. zscore = (data - mean) / stddev of x for each instrument within its group.
+========================= 操作符结束 =======================================
+
+========================= 数据字段开始 =======================================注意: DataField: 后面的是数据字段, DataFieldDescription: 此字段后面的是数据字段对应的描述或使用说明, DataFieldDescription字段后面的内容是使用说明, 不是数据字段
+
+DataField: call_breakeven_10
+DataFieldDescription: Price at which a stock's call options with expiration 10 days in the future break even based on its recent bid/ask mean.
+DataField: option_breakeven_180
+DataFieldDescription: Price at which a stock's options with expiration 180 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_90
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 90 days in the future.
+DataField: put_breakeven_270
+DataFieldDescription: Price at which a stock's put options with expiration 270 days in the future break even based on its recent bid/ask mean.
+DataField: put_breakeven_20
+DataFieldDescription: Price at which a stock's put options with expiration 20 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_all
+DataFieldDescription: Ratio of put volume to call volume for all maturities on stock's options.
+DataField: pcr_vol_180
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 180 days in the future.
+DataField: pcr_oi_150
+DataFieldDescription: Ratio of put open interest to call open interest on a stock's options with expiration 150 days in the future.
+DataField: forward_price_20
+DataFieldDescription: Forward price at 20 days derived from a synthetic long option with payoff similar to long stock + option dynamics. Combination of long ATM call and short ATM put.
+DataField: pcr_oi_90
+DataFieldDescription: Ratio of put open interest to call open interest on a stock's options with expiration 90 days in the future.
+DataField: pcr_oi_180
+DataFieldDescription: Ratio of put open interest to call open interest on a stock's options with expiration 180 days in the future.
+DataField: pcr_vol_360
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 360 days in the future.
+DataField: pcr_vol_270
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 270 days in the future.
+DataField: pcr_vol_20
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 20 days in the future.
+DataField: option_breakeven_10
+DataFieldDescription: Price at which a stock's options with expiration 10 days in the future break even based on its recent bid/ask mean.
+DataField: put_breakeven_30
+DataFieldDescription: Price at which a stock's put options with expiration 30 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_1080
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 1080 days in the future.
+DataField: call_breakeven_120
+DataFieldDescription: Price at which a stock's call options with expiration 120 days in the future break even based on its recent bid/ask mean.
+DataField: put_breakeven_360
+DataFieldDescription: Price at which a stock's put options with expiration 360 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_oi_30
+DataFieldDescription: Ratio of put open interest to call open interest on a stock's options with expiration 30 days in the future.
+DataField: pcr_oi_720
+DataFieldDescription: Ratio of put open interest to call open interest on a stock's options with expiration 720 days in the future.
+DataField: forward_price_90
+DataFieldDescription: Forward price at 90 days derived from a synthetic long option with payoff similar to long stock + option dynamics. Combination of long ATM call and short ATM put.
+DataField: pcr_vol_10
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 10 days in the future.
+DataField: option_breakeven_270
+DataFieldDescription: Price at which a stock's options with expiration 270 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_720
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 720 days in the future.
+DataField: call_breakeven_60
+DataFieldDescription: Price at which a stock's call options with expiration 60 days in the future break even based on its recent bid/ask mean.
+DataField: option_breakeven_150
+DataFieldDescription: Price at which a stock's options with expiration 150 days in the future break even based on its recent bid/ask mean.
+DataField: put_breakeven_90
+DataFieldDescription: Price at which a stock's put options with expiration 90 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_oi_all
+DataFieldDescription: Ratio of put open interest to call open interest for all maturities on stock's options.
+DataField: pcr_vol_30
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 30 days in the future.
+DataField: fnd6_newa2v1300_xopteps
+DataFieldDescription: Implied Option EPS Basic
+DataField: fnd6_newqeventv110_dilavq
+DataFieldDescription: Dilution Available - Excluding Extraordinary Items
+DataField: fnd6_newqeventv110_pnciaq
+DataFieldDescription: Core Pension Interest Adjustment After-tax
+DataField: fnd6_newqeventv110_ancq
+DataFieldDescription: Non-Current Assets - Total
+DataField: fnd6_newqeventv110_spiopq
+DataFieldDescription: Other Special Items Pretax
+DataField: fnd6_ppeveb
+DataFieldDescription: Property, Plant, and Equipment - Ending Balance (Schedule V)
+DataField: fnd6_newqv1300_optfvgrq
+DataFieldDescription: Options - Fair Value of Options Granted
+DataField: fnd6_optex
+DataFieldDescription: Options Exercisable (000)
+DataField: fnd6_newqv1300_xoptepsq
+DataFieldDescription: Implied Option EPS Basic
+DataField: fnd6_niadj
+DataFieldDescription: Net Income Adjusted for Common/Ordinary Stock (Capital) Equivalents
+DataField: fnd6_newqeventv110_prcpeps12
+DataFieldDescription: Core Post-Retirement Adjustment 12MM Basic EPS Effect Preliminary
+DataField: fnd6_pncdq
+DataFieldDescription: Core Pension Adjustment Diluted EPS Effect
+DataField: fnd6_recd
+DataFieldDescription: Receivables - Estimated Doubtful
+DataField: fnd6_newqeventv110_aocisecglq
+DataFieldDescription: Accum. Other Comp. Inc. - Unreal G/L Ret. Int. in Sec. Assets
+DataField: ebit
+DataFieldDescription: Earnings Before Interest and Taxes
+DataField: fnd6_acdo
+DataFieldDescription: Current Assets of Discontinued Operations
+DataField: fnd6_newqv1300_cicurrq
+DataFieldDescription: Comp Inc - Currency Trans Adj
+DataField: fnd6_incorp
+DataFieldDescription: Incorporated
+DataField: fnd6_newa1v1300_dltt
+DataFieldDescription: Long-Term Debt - Total
+DataField: fnd6_am
+DataFieldDescription: Amortization of Intangibles
+DataField: fnd6_newqeventv110_rcaq
+DataFieldDescription: Restructuring Cost After-tax
+DataField: fnd6_intc
+DataFieldDescription: Interest Capitalized
+DataField: fnd6_eventv110_gdwliepsq
+DataFieldDescription: Impairment of Goodwill Basic EPS Effect
+DataField: fnd6_aqi
+DataFieldDescription: Acquisitions - Income Contribution
+DataField: fnd6_cptnewqeventv110_apq
+DataFieldDescription: Accounts Payable/Creditors - Trade
+DataField: fnd6_newa2v1300_tstkn
+DataFieldDescription: Treasury Stock - Number of Common Shares
+DataField: fnd6_newqeventv110_lqpl1q
+DataFieldDescription: Liabilities Level 1 (Quoted Prices)
+DataField: fnd6_dcvt
+DataFieldDescription: Debt - Convertible
+DataField: fnd6_eventv110_wddq
+DataFieldDescription: Writedowns Diluted EPS Effect
+DataField: fnd6_newqeventv110_setd12
+DataFieldDescription: Settlement (Litigation/Insurance) Diluted EPS Effect 12MM
+DataField: scl12_alltype_buzzvec
+DataFieldDescription: sentiment volume
+DataField: scl12_alltype_sentvec
+DataFieldDescription: sentiment
+DataField: scl12_alltype_typevec
+DataFieldDescription: instrument type index
+DataField: scl12_buzz
+DataFieldDescription: relative sentiment volume
+DataField: scl12_buzz_fast_d1
+DataFieldDescription: relative sentiment volume
+DataField: scl12_buzzvec
+DataFieldDescription: sentiment volume
+DataField: scl12_sentiment
+DataFieldDescription: sentiment
+DataField: scl12_sentiment_fast_d1
+DataFieldDescription: sentiment
+DataField: scl12_sentvec
+DataFieldDescription: sentiment
+DataField: scl12_typevec
+DataFieldDescription: instrument type index
+DataField: snt_buzz
+DataFieldDescription: negative relative sentiment volume, fill nan with 0
+DataField: snt_buzz_bfl
+DataFieldDescription: negative relative sentiment volume, fill nan with 1
+DataField: snt_buzz_bfl_fast_d1
+DataFieldDescription: negative relative sentiment volume, fill nan with 1
+DataField: snt_buzz_fast_d1
+DataFieldDescription: negative relative sentiment volume, fill nan with 0
+DataField: snt_buzz_ret
+DataFieldDescription: negative return of relative sentiment volume
+DataField: snt_buzz_ret_fast_d1
+DataFieldDescription: negative return of relative sentiment volume
+DataField: snt_value
+DataFieldDescription: negative sentiment, fill nan with 0
+DataField: snt_value_fast_d1
+DataFieldDescription: negative sentiment, fill nan with 0
+DataField: analyst_revision_rank_derivative
+DataFieldDescription: Change in ranking for analyst revisions and momentum compared to previous period.
+DataField: cashflow_efficiency_rank_derivative
+DataFieldDescription: Change in ranking for cash flow generation and profitability compared to previous period.
+DataField: composite_factor_score_derivative
+DataFieldDescription: Change in overall composite factor score from the prior period.
+DataField: earnings_certainty_rank_derivative
+DataFieldDescription: Change in ranking for earnings sustainability and certainty compared to previous period.
+DataField: fscore_bfl_growth
+DataFieldDescription: The purpose of this metric is to qualify the expected MT growth potential of the stock.
+DataField: fscore_bfl_momentum
+DataFieldDescription: The purpose of this metric is to identify stocks which are currently undergoing either up or downward analyst revisions.
+DataField: fscore_bfl_profitability
+DataFieldDescription: The purpose of this metric is to rank stock based on their ability to generate cash flows.
+DataField: fscore_bfl_quality
+DataFieldDescription: The purpose of this metric is to measure both the sustainability and certainty of earnings.
+DataField: fscore_bfl_surface
+DataFieldDescription: The static score. An index between 0 & 100 is applied for each stock and each composite factor - The first ranking is a pentagon surface-based score. The larger the surface, the higher the rank.
+DataField: fscore_bfl_surface_accel
+DataFieldDescription: The derivative score. In a second step, we calculate the derivative of this score (ie: Is the surface of the pentagon increasing or decreasing from the previous month?).
+DataField: fscore_bfl_total
+DataFieldDescription: The final score M-Score is a weighted average of both the Pentagon surface score and the Pentagon acceleration score.
+DataField: fscore_bfl_value
+DataFieldDescription: The purpose of this metric is to see if the stock is under or overpriced given several well known valuation standards.
+DataField: fscore_growth
+DataFieldDescription: The purpose of this metric is to qualify the expected MT growth potential of the stock.
+DataField: fscore_momentum
+DataFieldDescription: The purpose of this metric is to identify stocks which are currently undergoing either up or downward analyst revisions.
+DataField: fscore_profitability
+DataFieldDescription: The purpose of this metric is to rank stock based on their ability to generate cash flows.
+DataField: fscore_quality
+DataFieldDescription: The purpose of this metric is to measure both the sustainability and certainty of earnings.
+DataField: fscore_surface
+DataFieldDescription: The static score. An index between 0 & 100 is applied for each stock and each composite factor - The first ranking is a pentagon surface-based score. The larger the surface, the higher the rank.
+DataField: fscore_surface_accel
+DataFieldDescription: The derivative score. In a second step, we calculate the derivative of this score (ie: Is the surface of the pentagon increasing or decreasing from the previous month?).
+DataField: fscore_total
+DataFieldDescription: The final score M-Score is a weighted average of both the Pentagon surface score and the Pentagon acceleration score.
+DataField: fscore_value
+DataFieldDescription: The purpose of this metric is to see if the stock is under or overpriced given several well known valuation standards.
+DataField: growth_potential_rank_derivative
+DataFieldDescription: Change in ranking for medium-term growth potential compared to previous period.
+DataField: multi_factor_acceleration_score_derivative
+DataFieldDescription: Change in the acceleration of multi-factor score compared to previous period.
+DataField: multi_factor_static_score_derivative
+DataFieldDescription: Change in static multi-factor score compared to previous period.
+DataField: relative_valuation_rank_derivative
+DataFieldDescription: Change in ranking for valuation metrics compared to previous period.
+DataField: snt_social_value
+DataFieldDescription: Z score of sentiment
+DataField: snt_social_volume
+DataFieldDescription: Normalized tweet volume
+DataField: beta_last_30_days_spy
+DataFieldDescription: Beta to SPY in 30 Days
+DataField: beta_last_360_days_spy
+DataFieldDescription: Beta to SPY in 360 Days
+DataField: beta_last_60_days_spy
+DataFieldDescription: Beta to SPY in 60 Days
+DataField: beta_last_90_days_spy
+DataFieldDescription: Beta to SPY in 90 Days
+DataField: correlation_last_30_days_spy
+DataFieldDescription: Correlation to SPY in 30 Days
+DataField: correlation_last_360_days_spy
+DataFieldDescription: Correlation to SPY in 360 Days
+DataField: correlation_last_60_days_spy
+DataFieldDescription: Correlation to SPY in 60 Days
+DataField: correlation_last_90_days_spy
+DataFieldDescription: Correlation to SPY in 90 Days
+DataField: systematic_risk_last_30_days
+DataFieldDescription: Systematic Risk Last 30 Days
+DataField: systematic_risk_last_360_days
+DataFieldDescription: Systematic Risk Last 360 Days
+DataField: systematic_risk_last_60_days
+DataFieldDescription: Systematic Risk Last 60 Days
+DataField: systematic_risk_last_90_days
+DataFieldDescription: Systematic Risk Last 90 Days
+DataField: unsystematic_risk_last_30_days
+DataFieldDescription: Unsystematic Risk Last 30 Days - Relative to SPY
+DataField: unsystematic_risk_last_360_days
+DataFieldDescription: Unsystematic Risk Last 360 Days - Relative to SPY
+DataField: unsystematic_risk_last_60_days
+DataFieldDescription: Unsystematic Risk Last 60 Days - Relative to SPY
+DataField: unsystematic_risk_last_90_days
+DataFieldDescription: Unsystematic Risk Last 90 Days - Relative to SPY
+DataField: operating_profit_before_interest_tax
+DataFieldDescription: Earnings Before Interest and Taxes (EBIT) - Actual Value
+DataField: anl4_cfi_high
+DataFieldDescription: Cash Flow From Investing - The highest estimation
+DataField: anl4_dez1basicafv4v104_est
+DataFieldDescription: Estimation value
+DataField: highest_sales_estimate
+DataFieldDescription: Sales - The highest estimation for the annual period
+DataField: anl4_fcf_high
+DataFieldDescription: Free cash flow - aggregation on estimations, max
+DataField: anl4_basicconltv110_mean
+DataFieldDescription: Mean of estimations
+DataField: anl4_adxqfv110_mean
+DataFieldDescription: Mean of estimations
+DataField: anl4_eaz1lqfv110_bk
+DataFieldDescription: Broker name (int)
+DataField: anl4_rd_exp_flag
+DataFieldDescription: Research and Development Expense - forecast type (revision/new/...)
+DataField: free_cash_flow_total
+DataFieldDescription: Free Cash Flow value - Annual
+DataField: min_pretax_profit_guidance
+DataFieldDescription: Minimum guidance value for Pretax income
+DataField: max_shareholders_equity_guidance
+DataFieldDescription: The maximum guidance value for Total Shareholders' Equity.
+DataField: est_netdebt
+DataFieldDescription: Net debt - mean of estimations
+DataField: anl4_fcf_number
+DataFieldDescription: Free Cash Flow - number of estimations
+DataField: eps_previous_estimate_value
+DataFieldDescription: The previous estimation of Earnings Per Share
+DataField: est_epsr
+DataFieldDescription: GAAP Earnings per share - mean of estimations
+DataField: anl4_fsguidanceqfv4_item
+DataFieldDescription: Financial item
+DataField: min_basic_shares_guidance
+DataFieldDescription: Shares Basic - Minimum guidance value
+DataField: anl4_bac1detailrec_item
+DataFieldDescription: Financial item
+DataField: net_debt_actual_value
+DataFieldDescription: Net debt- announced financial value
+DataField: anl4_qf_az_hgih_spe
+DataFieldDescription: Earnings per share - The highest estimation
+DataField: anl4_detailrecv4_est
+DataFieldDescription: Estimation value for recommendation detail
+DataField: anl4_qfv4_eps_low
+DataFieldDescription: Earnings per share - The lowest estimation
+DataField: anl4_basicconqfv110_low
+DataFieldDescription: The lowest estimation
+DataField: anl4_afv4_eps_number
+DataFieldDescription: Earnings per share - number of estimations for annual frequency
+DataField: earnings_per_share_standard_deviation
+DataFieldDescription: Earnings per share - standard deviation of estimations
+DataField: anl4_ptpr_number
+DataFieldDescription: Reported Pretax Income - number of estimations
+DataField: anl4_gric_number
+DataFieldDescription: Gross income - number of estimations
+DataField: pretax_income_total
+DataFieldDescription: Pretax Profit - Value for the annual period
+DataField: ebit_reported_value
+DataFieldDescription: Earnings Before Interest & Taxes - actual value for the quarter
+DataField: pv13_hierarchy_min51_f4_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min10_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min5_f3g2_sector
+DataFieldDescription: grouping fields
+DataField: pv13_com_page_rank
+DataFieldDescription: the PageRank of competitors
+DataField: pv13_r2_min20_1000_sector
+DataFieldDescription: grouping fields
+DataField: pv13_revere_zipcode
+DataFieldDescription: Zip code
+DataField: pv13_custretsig_retsig
+DataFieldDescription: Sign of customer return
+DataField: pv13_h_min24_500_sector
+DataFieldDescription: Grouping fields for top 500
+DataField: pv13_hierarchy_min5_1000_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min20_top3000_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_2l_scibr
+DataFieldDescription: grouping fields
+DataField: rel_ret_all
+DataFieldDescription: Averaged one-day return of the companies whose product overlapped with the instrument
+DataField: pv13_rha2_min5_sector
+DataFieldDescription: grouping fields
+DataField: pv13_h_min30_3000_mapped_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min30_3000_mapped_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchys32_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_f2_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_new_6l_scibr
+DataFieldDescription: grouping fields
+DataField: rel_num_part
+DataFieldDescription: number of the instrument's partners
+DataField: pv13_hierarchy_min2_focused_pureplay_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min51_f2_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min10_industry_3000_sector
+DataFieldDescription: grouping fields
+DataField: pv13_rha2_min10_sector
+DataFieldDescription: grouping fields
+DataField: pv13_rcsed_6l
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min5_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min2_pureplay_only_sector
+DataFieldDescription: grouping fields
+DataField: pv13_new_3l_scibr
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min20_3k_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min40_3000_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_rha2_min30_3000_513_sector
+DataFieldDescription: grouping fields
+DataField: implied_volatility_call_90
+DataFieldDescription: At-the-money option-implied volatility for call Option for 90 days
+DataField: historical_volatility_150
+DataFieldDescription: Close-to-close Historical volatility over 150 days
+DataField: implied_volatility_call_1080
+DataFieldDescription: At-the-money option-implied volatility for call option for 1080 days
+DataField: implied_volatility_mean_20
+DataFieldDescription: At-the-money option-implied volatility mean for 20 days
+DataField: implied_volatility_mean_skew_120
+DataFieldDescription: At-the-money option-implied volatility mean skew for 120 days
+DataField: historical_volatility_90
+DataFieldDescription: Close-to-close Historical volatility over 90 days
+DataField: implied_volatility_call_150
+DataFieldDescription: At-the-money option-implied volatility for call Option for 150 days
+DataField: implied_volatility_mean_skew_270
+DataFieldDescription: At-the-money option-implied volatility mean skew for 270 days
+DataField: implied_volatility_put_1080
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 3 years
+DataField: parkinson_volatility_90
+DataFieldDescription: Parkinson model's historical volatility over 90 days
+DataField: implied_volatility_put_60
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 60 days
+DataField: implied_volatility_put_150
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 150 days
+DataField: implied_volatility_mean_180
+DataFieldDescription: At-the-money option-implied volatility mean for 180 days
+DataField: historical_volatility_10
+DataFieldDescription: Close-to-close Historical volatility over 10 days
+DataField: implied_volatility_put_20
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 20 days
+DataField: historical_volatility_60
+DataFieldDescription: Close-to-close Historical volatility over 60 days
+DataField: implied_volatility_call_270
+DataFieldDescription: At-the-money option-implied volatility for call Option for 270 days
+DataField: implied_volatility_mean_skew_150
+DataFieldDescription: At-the-money option-implied volatility mean skew for 150 days
+DataField: implied_volatility_mean_skew_180
+DataFieldDescription: At-the-money option-implied volatility mean skew for 180 days
+DataField: implied_volatility_put_30
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 30 days
+DataField: implied_volatility_call_10
+DataFieldDescription: At-the-money option-implied volatility for call Option for 10 days
+DataField: parkinson_volatility_30
+DataFieldDescription: Parkinson model's historical volatility over 30 days
+DataField: parkinson_volatility_60
+DataFieldDescription: Parkinson model's historical volatility over 60 days
+DataField: historical_volatility_180
+DataFieldDescription: Close-to-close Historical volatility over 180 days
+DataField: parkinson_volatility_120
+DataFieldDescription: Parkinson model's historical volatility over 120 days
+DataField: implied_volatility_mean_skew_10
+DataFieldDescription: At-the-money option-implied volatility mean skew for 10 days
+DataField: implied_volatility_call_720
+DataFieldDescription: At-the-money option-implied volatility for call Option for 720 days
+DataField: parkinson_volatility_180
+DataFieldDescription: Parkinson model's historical volatility over 180 days
+DataField: implied_volatility_put_360
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 360 days
+DataField: implied_volatility_put_720
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 720 days
+DataField: nws12_afterhsz_maxup
+DataFieldDescription: Percent change from the price at the time of the news to the after the news high
+DataField: nws12_mainz_5l
+DataFieldDescription: Number of minutes that elapsed before price went up 5 percentage points
+DataField: news_pct_120min
+DataFieldDescription: The percent change in price in the first 120 minutes following the news release
+DataField: nws12_afterhsz_spylast
+DataFieldDescription: Last Price of the SPY at the time of the news
+DataField: nws12_prez_mainvwap
+DataFieldDescription: Main session volume-weighted average price
+DataField: news_curr_vol
+DataFieldDescription: Current day's session volume
+DataField: news_prev_vol
+DataFieldDescription: Previous day's session volume
+DataField: news_eod_high
+DataFieldDescription: Highest price reached between the time of news and the end of the session
+DataField: nws12_afterhsz_short_interest
+DataFieldDescription: Total number of shares sold short divided by total number of shares outstanding
+DataField: nws12_prez_02l
+DataFieldDescription: Number of minutes that elapsed before price went up 20 percentage points
+DataField: nws12_mainz_5p
+DataFieldDescription: The minimum of L or S above for 5-minute bucket
+DataField: nws12_prez_atrratio
+DataFieldDescription: Ratio of Today Range to 20-day average true range
+DataField: nws12_prez_prev_vol
+DataFieldDescription: Previous day's session volume
+DataField: nws12_prez_peratio
+DataFieldDescription: Reported price to earnings ratio for the calendar day of the session
+DataField: nws12_prez_mktcap
+DataFieldDescription: Reported market capitalization for the calendar day of the session
+DataField: nws12_afterhsz_5l
+DataFieldDescription: Number of minutes that elapsed before price went up 5 percentage points
+DataField: nws12_mainz_vol_ratio
+DataFieldDescription: Curr_Vol / Mov_Vol
+DataField: nws12_afterhsz_highexcstddev
+DataFieldDescription: (EODHigh - TONLast)/StdDev, where StdDev is one standard deviation for the close price for 30 calendar days
+DataField: nws12_afterhsz_1l
+DataFieldDescription: Number of minutes that elapsed before price went up 1 percentage points
+DataField: nws12_afterhsz_01l
+DataFieldDescription: Number of minutes that elapsed before price went up 10 percentage points
+DataField: nws12_mainz_prevwap
+DataFieldDescription: Pre session volume weighted average price
+DataField: news_cap
+DataFieldDescription: Reported market capitalization for the calendar day of the session
+DataField: nws12_afterhsz_newrecord
+DataFieldDescription: Tracks whether the news is first instance or a duplicate
+DataField: nws12_afterhsz_tonlast
+DataFieldDescription: Price at the time of news
+DataField: news_close_vol
+DataFieldDescription: Main close volume
+DataField: nws12_afterhsz_1s
+DataFieldDescription: Number of minutes that elapsed before price went down 1 percentage point
+DataField: nws12_mainz_postvwap
+DataFieldDescription: Post session volume weighted average price
+DataField: nws12_mainz_mainvwap
+DataFieldDescription: Main session volume weighted average price
+DataField: news_mins_10_chg
+DataFieldDescription: The minimum of L or S above for 10-minute bucket
+DataField: nws12_mainz_maxdown
+DataFieldDescription: Percent change from the price at the time of the news to the after-the-news low
+DataField: top1000
+DataFieldDescription: 20140630
+DataField: top200
+DataFieldDescription: 20140630
+DataField: top3000
+DataFieldDescription: 20140630
+DataField: top500
+DataFieldDescription: 20140630
+DataField: topsp500
+DataFieldDescription: 20140630
+DataField: rp_nip_partner
+DataFieldDescription: News impact projection of partnership news
+DataField: rp_nip_technical
+DataFieldDescription: News impact projection based on technical analysis
+DataField: nws18_qcm
+DataFieldDescription: News sentiment of relevant news with high confidence
+DataField: nws18_bee
+DataFieldDescription: News sentiment specializing in growth of earnings
+DataField: rp_css_product
+DataFieldDescription: Composite sentiment score of product and service-related news
+DataField: rp_ess_product
+DataFieldDescription: Event sentiment score of product and service-related news
+DataField: rp_css_inverstor
+DataFieldDescription: Composite sentiment score of investor relations news
+DataField: nws18_qmb
+DataFieldDescription: News sentiment specializing in editorials on global markets
+DataField: rp_nip_legal
+DataFieldDescription: News impact projection of legal news
+DataField: rp_ess_business
+DataFieldDescription: Event sentiment score of business-related news
+DataField: rp_nip_ptg
+DataFieldDescription: News impact projection of price target news
+DataField: rp_nip_price
+DataFieldDescription: News impact projection of stock price news
+DataField: nws18_ber
+DataFieldDescription: News sentiment specializing in earnings result
+DataField: nws18_relevance
+DataFieldDescription: Relevance of news to the company
+DataField: rp_ess_equity
+DataFieldDescription: Event sentiment score of equity action news
+DataField: rp_ess_assets
+DataFieldDescription: Event sentiment score of assets news
+DataField: rp_nip_revenue
+DataFieldDescription: News impact projection of revenue news
+DataField: nws18_acb
+DataFieldDescription: News sentiment specializing in corporate action announcements
+DataField: rp_ess_earnings
+DataFieldDescription: Event sentiment score of earnings news
+DataField: rp_ess_ratings
+DataFieldDescription: Event sentiment score of analyst ratings-related news
+DataField: rp_nip_inverstor
+DataFieldDescription: News impact projection of investor relations news
+DataField: rp_css_society
+DataFieldDescription: Composite sentiment score of society-related news
+DataField: rp_ess_partner
+DataFieldDescription: Event sentiment score of partnership news
+DataField: rp_nip_earnings
+DataFieldDescription: News impact projection of earnings news
+DataField: rp_css_ratings
+DataFieldDescription: Composite sentiment score of analyst ratings-related news
+DataField: rp_nip_credit
+DataFieldDescription: News impact projection of credit news
+DataField: nws18_sse
+DataFieldDescription: Sentiment of phrases impacting the company
+DataField: rp_css_dividends
+DataFieldDescription: Composite sentiment score of dividends news
+DataField: rp_nip_equity
+DataFieldDescription: News impact projection of equity action news
+DataField: rp_ess_price
+DataFieldDescription: Event sentiment score of stock price news
+DataField: fn_repayments_of_lines_of_credit_a
+DataFieldDescription: Amount of cash outflow for payment of an obligation from a lender, including but not limited to, letter of credit, standby letter of credit and revolving credit arrangements.
+DataField: fn_unrecognized_tax_benefits_a
+DataFieldDescription: Amount of unrecognized tax benefits.
+DataField: fn_oth_income_loss_derivatives_qualifying_as_hedges_of_tax_q
+DataFieldDescription: Amount after tax and reclassification adjustments, of increase (decrease) in accumulated gain (loss) from derivative instruments designated and qualifying as the effective portion of cash flow hedges and an entity's share of an equity investee's increase (decrease) in deferred hedging gain (loss).
+DataField: fn_comp_number_of_shares_authorized_a
+DataFieldDescription: Count of unique IDs of industry participants. Industry stands for an aggregate view of all  equity clearance activity for the date, symbol, and transaction type in question.
+DataField: fn_finite_lived_intangible_assets_net_q
+DataFieldDescription: Finite Lived Intangible Assets, Net
+DataField: fnd2_a_ltrmdmrepoplinytwo
+DataFieldDescription: Amount of long-term debt payable, sinking fund requirements, and other securities issued that are redeemable by holder at fixed or determinable prices and dates maturing in the 2nd fiscal year following the latest fiscal year. Excludes interim and annual periods when interim periods are reported on a rolling approach, from latest balance sheet date.
+DataField: fn_amortization_of_intangible_assets_q
+DataFieldDescription: The aggregate expense charged against earnings to allocate the cost of intangible assets (nonphysical assets not used in production) in a systematic and rational manner to the periods expected to benefit from such assets. As a noncash expense, this element is added back to net income when calculating cash provided by or used in operations using the indirect method.
+DataField: fnd2_a_inventoryrawmaterials
+DataFieldDescription: Amount before valuation and LIFO reserves of raw materials expected to be sold, or consumed within 1 year or operating cycle, if longer.
+DataField: fn_mne_a
+DataFieldDescription: Amount before accumulated depreciation of tangible personal property used to produce goods and services, including, but is not limited to, tools, dies and molds, computer and office equipment.
+DataField: fn_oth_income_loss_available_for_sale_securities_adj_of_tax_q
+DataFieldDescription: Amount after tax and reclassification adjustments, of appreciation (loss) in value of unsold available-for-sale securities. Excludes amounts related to other than temporary impairment (OTTI) loss.
+DataField: fn_derivative_fair_value_of_derivative_liability_a
+DataFieldDescription: Fair value, before effects of master netting arrangements, of a financial liability or contract with one or more underlyings, notional amount or payment provision or both, and the contract can be net settled by means outside the contract or delivery of an asset. Includes liabilities elected not to be offset. Excludes liabilities not subject to a master netting arrangement.
+DataField: fnd2_dbplanepdfbnfpy5
+DataFieldDescription: Amount of benefits from a defined benefit plan expected to be paid in the 5th fiscal year following the latest fiscal year. Excludes interim and annual periods when interim periods are reported on a rolling approach, from latest balance sheet date.
+DataField: fnd2_dfdtxasoprlcarryfwd
+DataFieldDescription: Amount before allocation of valuation allowances of deferred tax asset attributable to deductible operating loss carryforwards.
+DataField: fn_finite_lived_intangible_assets_net_a
+DataFieldDescription: Finite Lived Intangible Assets, Net
+DataField: fnd2_a_sbcpnargmtwfsptepddvdrt
+DataFieldDescription: The estimated dividend rate (a percentage of the share price) to be paid (expected dividends) to holders of the underlying shares over the option's term.
+DataField: fn_comp_not_rec_q
+DataFieldDescription: Unrecognized cost of unvested share-based compensation awards.
+DataField: fn_debt_instrument_face_amount_q
+DataFieldDescription: Debt face amount
+DataField: fnd2_dbplanepdfbnfpnext12m
+DataFieldDescription: Amount of benefits from a defined benefit plan expected to be paid in the next fiscal year following the latest fiscal year. Excludes interim and annual periods when interim periods are reported on a rolling approach, from latest balance sheet date.
+DataField: fnd2_a_flintasamt1expythree
+DataFieldDescription: Amount of amortization expense for assets, excluding financial assets and goodwill, lacking physical substance with a finite life expected to be recognized during the 3rd fiscal year following the latest fiscal year. Excludes interim and annual periods when interim periods are reported on a rolling approach, from latest balance sheet date.
+DataField: fn_proceeds_from_issuance_of_debt_a
+DataFieldDescription: The cash inflow during the period from additional borrowings in aggregate debt. Includes proceeds from short-term and long-term debt.
+DataField: fnd2_dfdlocalitxexp
+DataFieldDescription: Income Tax Expense, Deferred
+DataField: fnd2_a_lineofcrfcyrmbrgcap
+DataFieldDescription: Amount of borrowing capacity currently available under the credit facility (current borrowing capacity less the amount of borrowings outstanding).
+DataField: fnd2_a_sbcpnargmpmtwgtm
+DataFieldDescription: The weighted average period between the balance sheet date and expiration for all awards outstanding under the plan, which may be expressed in a decimal value for number of years.
+DataField: fn_interest_paid_net_q
+DataFieldDescription: Net interest
+DataField: fnd2_dfdtxlbspropplteqm
+DataFieldDescription: Amount of deferred tax liability attributable to taxable temporary differences from property, plant, and equipment.
+DataField: fn_derivative_fair_value_of_derivative_liability_q
+DataFieldDescription: Fair value, before effects of master netting arrangements, of a financial liability or contract with one or more underlyings, notional amount or payment provision or both, and the contract can be net settled by means outside the contract or delivery of an asset. Includes liabilities elected not to be offset. Excludes liabilities not subject to a master netting arrangement.
+DataField: fnd2_q_seniornotes
+DataFieldDescription: Including the current and noncurrent portions, carrying value as of the balance sheet date of Notes with the highest claim on the assets of the issuer in case of bankruptcy or liquidation (with maturities initially due after 1 year or beyond the operating cycle if longer). Senior note holders are paid off in full before any payments are made to junior note holders.
+DataField: fn_line_of_credit_facility_amount_out_q
+DataFieldDescription: Amount borrowed under the credit facility as of the balance sheet date.
+DataField: fn_prepaid_expense_q
+DataFieldDescription: Carrying amount for an unclassified balance sheet date of expenditures made in advance of when the economic benefit of the cost will be realized, and which will be expensed in future periods with the passage of time or when a triggering event occurs. For a classified balance sheet, represents the noncurrent portion of prepaid expenses (the current portion has a separate concept).
+DataField: fn_finite_lived_intangible_assets_acq_q
+DataFieldDescription: Amount of assets, excluding financial assets and goodwill, lacking physical substance with a finite life acquired.
+DataField: adv20
+DataFieldDescription: Average daily volume in past 20 days
+DataField: cap
+DataFieldDescription: Daily market capitalization (in millions)
+DataField: close
+DataFieldDescription: Daily close price
+DataField: country
+DataFieldDescription: Country grouping
+DataField: currency
+DataFieldDescription: Currency
+DataField: cusip
+DataFieldDescription: CUSIP Value
+DataField: dividend
+DataFieldDescription: Dividend
+DataField: exchange
+DataFieldDescription: Exchange grouping
+DataField: high
+DataFieldDescription: Daily high price
+DataField: industry
+DataFieldDescription: Industry grouping
+DataField: isin
+DataFieldDescription: ISIN Value
+DataField: low
+DataFieldDescription: Daily low price
+DataField: market
+DataFieldDescription: Market grouping
+DataField: open
+DataFieldDescription: Daily open price
+DataField: returns
+DataFieldDescription: Daily returns
+DataField: sector
+DataFieldDescription: Sector grouping
+DataField: sedol
+DataFieldDescription: Sedol
+DataField: sharesout
+DataFieldDescription: Daily outstanding shares (in millions)
+DataField: split
+DataFieldDescription: Stock split ratio
+DataField: subindustry
+DataFieldDescription: Subindustry grouping
+DataField: ticker
+DataFieldDescription: Ticker
+DataField: volume
+DataFieldDescription: Daily volume
+DataField: vwap
+DataFieldDescription: Daily volume weighted average price
+========================= 数据字段结束 =======================================
+

manual_prompt/manual_prompt_20251216105934.txt

@@ -0,0 +1,891 @@
+任务指令
+你是一个WorldQuant WebSim因子工程师。你的任务是生成 20 个用于行业轮动策略的复合型Alpha因子表达式。
+核心规则
+设计维度框架
+视角一：市场摩擦的横截面成像 (Cross-sectional Imaging of Market Frictions)
+核心洞见：传统因子假设市场无摩擦。真正的Alpha可能藏于“摩擦本身”——即指令流（order flow）转化为价格运动时，在不同股票上表现出的效率差异图谱。我们不为消除摩擦建模，而是主动测绘它。
+研究提示词 (用于指导因子构建)：
+指令流冲击的“消化速率”：测量单位异常交易量（可定义为其z-score）所引发的瞬时价格冲击（如未来1-5分钟收益率），以及该冲击在随后较短时间（如30分钟、1小时）的衰减半衰期。寻找“冲击大但衰减慢”（高摩擦、低流动性）与“冲击小但衰减快”（低摩擦、高流动性）的股票，并研究其横截面收益预测能力。
+买卖失衡的“路径依赖”：考察买单流与卖单流（可用正负交易额近似）的净额在时间序列上的自相关性模式。是呈现“均值回归”（失衡迅速被套利消化）还是“趋势持续”（失衡自我强化）？这种模式在不同波动率、不同市值股票中如何变化？能否构建一个度量“订单流趋势持续性”的指标？
+价格发现的“领地性”：将价格变化分解为“由自身交易驱动”的部分和“由同行业龙头/指数驱动”的部分（可通过日内收益率与龙头/指数收益率的滚动协方差分解）。计算“自身解释比例”。该比例高的股票，其价格发现是“内生性”的；比例低的则是“外生性”或“跟随性”的。这两类股票在不同市场环境（牛市/熊市/震荡市）下的表现有何系统性差异？
+视角二：投资者注意力的生态学建模 (Ecology of Investor Attention)
+核心洞见：注意力是金融市场最稀缺的资源。市场不是信息的聚合器，而是注意力资源的分配系统。因子应捕捉注意力的“聚集-分散”、“转移-停滞”动态，而非信息本身。
+研究提示词 (用于指导因子构建)：
+注意力“聚焦度”与“涣散度”：用交易量在时间序列上的分布熵来衡量。例如，过去20个交易日内，交易量分布的基尼系数或赫芬达尔指数。高集中度（某几天放巨量）意味着注意力短期爆发；低集中度（量能均匀）意味着持续平淡的关注。研究这两种状态转换前后，股票的收益特征。
+行业内注意力的“捕食-被捕食”关系：定义“注意力领导者”（如当日涨幅前3的股票）和“注意力追随者”（同行业其他股票）。计算领导股出现后，追随者的成交量与价格在随后N个时间段内的响应速度和强度。这种响应的不对称性（有些股票迅速响应，有些滞后）能否预测未来的相对强弱？
+“注意力惯性”与“注意力反转”：度量一个股票在失去短期催化剂（如公告、事件）后，其异常成交量（或搜索量、社交媒体活跃度）回落至基线水平的速度。回落慢的股票具有“注意力惯性”，可能存在定价偏差的持续；回落快的股票则“注意力反转”迅速。构建一个“注意力衰减时间”因子。
+视角三：价格运动的“形态语法”与“叙事连贯性” (Morphological Syntax & Narrative Coherence of Price Movements)
+核心洞见：价格运动不仅是有方向的，更是有“形状”和“语法”的。如同语言，一段价格走势是否存在合“语法”的叙事结构（如“筑底-突破-回踩-主升”），还是杂乱无章的“噪音词汇”？市场参与者会潜意识地识别并交易这些“连贯叙事”。
+研究提示词 (用于指导因子构建)：
+K线序列的“可压缩性”：使用一种简化的算法（如趋势线拟合后的残差大小，或一定容忍度下的分段线性近似所需节点数），来衡量过去一段价格序列的信息复杂度。低复杂度（高可压缩性）意味着价格运动清晰、有规律；高复杂度则意味着混乱。研究“从混乱转向清晰”或“从清晰转向混乱”的拐点。
+关键价位互动的“叙事逻辑”：识别近期的显著高点、低点、缺口、密集成交区作为“关键叙事节点”。分析当前价格在接近这些节点时的行为：是“尊重”（反弹/受阻）还是“无视”（直接穿越）？一系列对历史节点的“尊重”行为构成了一个连贯的“支撑阻力叙事”。能否量化这种叙事逻辑的强度？
+多尺度运动的“相位同步”：选取短期（如5日）、中期（20日）、长期（60日）的价格滤波序列（如移动平均线）。计算它们之间方向变化的领先滞后关系与同步性（可类似相位角分析）。例如，短期均线是否总是率先转向并引导中长期？还是三者经常背离？“多周期共振向上/向下”这种高度同步的状态，其形成过程和瓦解过程有何预测意义？
+
+=== 关键语法规则（必须遵守） ===
+1. 数据字段规范：
+   - 可使用字段：close, volume, returns
+   - ❌ 错误：market_cap, marketcap, mkt_cap [这些字段不存在]
+   - ✅ 正确：使用volume作为规模代理，close作为价格
+   - returns通常定义为：ts_delta(close, 1) 或 close/ts_delay(close,1)-1
+2. ts_regression使用规范：
+   - 避免深度嵌套ts_regression，特别是作为其他函数的参数
+   - ✅ 正确：reg_slope = ts_regression(close, ts_step(1), 30, 0, 1)
+   - ❌ 错误：ts_delta(ts_regression(close, ts_step(1), 30, 0, 1), 5)
+   - 替代方案：用ts_delta组合计算动量变化
+3. if_else使用规范：
+   - 条件必须是简单布尔表达式
+   - 避免序列比较：❌ ts_std_dev(returns,60) > ts_mean(ts_std_dev(returns,60),120)
+   - 正确使用：✅ if_else(ts_rank(ts_std_dev(returns,60), 120) > 0.7, 短期动量, 长期动量)
+4. bucket函数使用规范：
+   - bucket()返回分组ID，可用于条件判断
+   - ✅ 正确：bucket(rank(volume), range="0,3,0.4") == 0 [第一组为大成交量]
+   - ✅ 正确：group_mean(x, 1, bucket(rank(volume), range="0,3,0.4"))
+   - 注意字符串格式：range="起始值,组数,步长" 或 buckets="分割点列表"
+=== 关键语法规则结束 ===
+*=====*
+注意事项:
+1. 避免过度复杂的嵌套（建议不超过3层）
+2. 每个表达式应有明确的经济逻辑
+3. 考虑实际交易可行性（避免未来函数）
+4. 包含风险控制元素（如波动率调整）
+5. 只能使用可用的数据字段：close, volume, returns等
+*=====*
+参数逻辑：参数d（回顾期）应在[5, 10, 20, 30, 60, 120]等具有市场意义（周、月、季度、半年）的数值中合理选择并差异化。
+行业隐含：通过group_mean、group_rank等函数或假设表达式在行业指数上运行来体现"行业"逻辑。
+构建框架指导（请按此逻辑创造新因子）：
+维度融合模板（选择至少2个）：
+A. 领导力动量 = 时序动量 × 横截面调整
+   逻辑：大成交量股票的动量更强
+   结构：group_mean(ts_delta(close, d1), 1, bucket(rank(volume), range="0,3,0.4"))
+B. 状态自适应动量 = 条件选择动量
+   逻辑：高波动用短期动量，低波动用长期动量
+   结构：if_else(ts_std_dev(returns,20) > 0.02, ts_delta(close,5), ts_delta(close,20))
+C. 行业传导因子 = 领先行业动量 × 相关性强度
+   逻辑：与强势行业相关性高的行业未来表现好
+   结构：multiply(ts_corr(group_mean(returns,1,industry), group_mean(returns,1,sector), d1), ts_delta(close,d2))
+D. 情绪反转 = 过度交易信号 × 基础趋势
+   逻辑：过度交易时反转，趋势延续时跟随
+   结构：multiply(reverse(ts_rank(volume/ts_mean(volume,20), 10)), ts_delta(close,20))
+关键组件库（可自由组合）：
+1. 动量类：ts_delta(close,{d}), ts_regression(close,ts_step(1),{d},0,1)
+2. 波动类：ts_std_dev(returns,{d}), ts_mean(abs(returns),{d})
+3. 成交量类：volume/ts_mean(volume,{d}), ts_zscore(volume,{d})
+4. 横截面类：if_else(rank(volume) > 阈值, 值1, 值2), bucket(rank(volume), range="0,3,0.4")
+5. 相关性类：ts_corr({x},{y},{d})
+6. 条件逻辑：if_else({condition}, {true_value}, {false_value})
+参数池：d ∈ [5,10,20,30,60,120], 阈值 ∈ [0.5,0.7,0.8]
+*=====*
+输出格式：
+输出必须是且仅是纯文本。
+每一行是一个完整、独立、语法正确的WebSim表达式。
+严禁任何形式的解释、编号、标点包裹（如引号）、Markdown格式或额外文本。
+===================== !!! 重点(输出方式) !!! =====================
+现在，请严格遵守以上所有规则，开始生成可立即在WebSim中运行的复合因子表达式。
+**输出格式**(一行一个表达式, 每个表达式中间需要添加一个空行, 只要表达式本身, 不要解释, 不需要序号, 也不要输出多余的东西)：
+表达式
+表达式
+表达式
+...
+表达式
+=================================================================
+重申：请确保所有表达式都使用WorldQuant WebSim平台函数，不要使用pandas、numpy或其他Python库函数。输出必须是一行有效的WQ表达式。
+以下是我的账号有权限使用的操作符, 请严格按照操作符, 进行生成,组合因子：
+
+以下是我的账号有权限使用的操作符, 请严格按照操作符, 进行生成,组合因子
+
+========================= 操作符开始 =======================================注意: Operator: 后面的是操作符,
+Description: 此字段后面的是操作符对应的描述或使用说明, Description字段后面的内容是使用说明, 不是操作符
+特别注意!!!! 必须按照操作符字段Operator的使用说明生成 alphaOperator: abs(x)
+Description: Absolute value of x
+Operator: add(x, y, filter = false)
+Description: Add all inputs (at least 2 inputs required). If filter = true, filter all input NaN to 0 before adding
+Operator: densify(x)
+Description: Converts a grouping field of many buckets into lesser number of only available buckets so as to make working with grouping fields computationally efficient
+Operator: divide(x, y)
+Description: x / y
+Operator: inverse(x)
+Description: 1 / x
+Operator: log(x)
+Description: Natural logarithm. For example: Log(high/low) uses natural logarithm of high/low ratio as stock weights.
+Operator: max(x, y, ..)
+Description: Maximum value of all inputs. At least 2 inputs are required
+Operator: min(x, y ..)
+Description: Minimum value of all inputs. At least 2 inputs are required
+Operator: multiply(x ,y, ... , filter=false)
+Description: Multiply all inputs. At least 2 inputs are required. Filter sets the NaN values to 1
+Operator: power(x, y)
+Description: x ^ y
+Operator: reverse(x)
+Description: - x
+Operator: sign(x)
+Description: if input > 0, return 1; if input < 0, return -1; if input = 0, return 0; if input = NaN, return NaN;
+Operator: signed_power(x, y)
+Description: x raised to the power of y such that final result preserves sign of x
+Operator: sqrt(x)
+Description: Square root of x
+Operator: subtract(x, y, filter=false)
+Description: x-y. If filter = true, filter all input NaN to 0 before subtracting
+Operator: and(input1, input2)
+Description: Logical AND operator, returns true if both operands are true and returns false otherwise
+Operator: if_else(input1, input2, input 3)
+Description: If input1 is true then return input2 else return input3.
+Operator: input1 < input2
+Description: If input1 < input2 return true, else return false
+Operator: input1 <= input2
+Description: Returns true if input1 <= input2, return false otherwise
+Operator: input1 == input2
+Description: Returns true if both inputs are same and returns false otherwise
+Operator: input1 > input2
+Description: Logic comparison operators to compares two inputs
+Operator: input1 >= input2
+Description: Returns true if input1 >= input2, return false otherwise
+Operator: input1!= input2
+Description: Returns true if both inputs are NOT the same and returns false otherwise
+Operator: is_nan(input)
+Description: If (input == NaN) return 1 else return 0
+Operator: not(x)
+Description: Returns the logical negation of x. If x is true (1), it returns false (0), and if input is false (0), it returns true (1).
+Operator: or(input1, input2)
+Description: Logical OR operator returns true if either or both inputs are true and returns false otherwise
+Operator: days_from_last_change(x)
+Description: Amount of days since last change of x
+Operator: hump(x, hump = 0.01)
+Description: Limits amount and magnitude of changes in input (thus reducing turnover)
+Operator: kth_element(x, d, k)
+Description: Returns K-th value of input by looking through lookback days. This operator can be used to backfill missing data if k=1
+Operator: last_diff_value(x, d)
+Description: Returns last x value not equal to current x value from last d days
+Operator: ts_arg_max(x, d)
+Description: Returns the relative index of the max value in the time series for the past d days. If the current day has the max value for the past d days, it returns 0. If previous day has the max value for the past d days, it returns 1
+Operator: ts_arg_min(x, d)
+Description: Returns the relative index of the min value in the time series for the past d days; If the current day has the min value for the past d days, it returns 0; If previous day has the min value for the past d days, it returns 1.
+Operator: ts_av_diff(x, d)
+Description: Returns x - tsmean(x, d), but deals with NaNs carefully. That is NaNs are ignored during mean computation
+Operator: ts_backfill(x,lookback = d, k=1, ignore="NAN")
+Description: Backfill is the process of replacing the NAN or 0 values by a meaningful value (i.e., a first non-NaN value)
+Operator: ts_corr(x, y, d)
+Description: Returns correlation of x and y for the past d days
+Operator: ts_count_nans(x ,d)
+Description: Returns the number of NaN values in x for the past d days
+Operator: ts_covariance(y, x, d)
+Description: Returns covariance of y and x for the past d days
+Operator: ts_decay_linear(x, d, dense = false)
+Description: Returns the linear decay on x for the past d days. Dense parameter=false means operator works in sparse mode and we treat NaN as 0. In dense mode we do not.
+Operator: ts_delay(x, d)
+Description: Returns x value d days ago
+Operator: ts_delta(x, d)
+Description: Returns x - ts_delay(x, d)
+Operator: ts_mean(x, d)
+Description: Returns average value of x for the past d days.
+Operator: ts_product(x, d)
+Description: Returns product of x for the past d days
+Operator: ts_quantile(x,d, driver="gaussian" )
+Description: It calculates ts_rank and apply to its value an inverse cumulative density function from driver distribution. Possible values of driver (optional ) are "gaussian", "uniform", "cauchy" distribution where "gaussian" is the default.
+Operator: ts_rank(x, d, constant = 0)
+Description: Rank the values of x for each instrument over the past d days, then return the rank of the current value + constant. If not specified, by default, constant = 0.
+Operator: ts_regression(y, x, d, lag = 0, rettype = 0)
+Description: Returns various parameters related to regression function
+Operator: ts_scale(x, d, constant = 0)
+Description: Returns (x - ts_min(x, d)) / (ts_max(x, d) - ts_min(x, d)) + constant. This operator is similar to scale down operator but acts in time series space
+Operator: ts_std_dev(x, d)
+Description: Returns standard deviation of x for the past d days
+Operator: ts_step(1)
+Description: Returns days' counter
+Operator: ts_sum(x, d)
+Description: Sum values of x for the past d days.
+Operator: ts_zscore(x, d)
+Description: Z-score is a numerical measurement that describes a value's relationship to the mean of a group of values. Z-score is measured in terms of standard deviations from the mean: (x - tsmean(x,d)) / tsstddev(x,d). This operator may help reduce outliers and drawdown.
+Operator: normalize(x, useStd = false, limit = 0.0)
+Description: Calculates the mean value of all valid alpha values for a certain date, then subtracts that mean from each element
+Operator: quantile(x, driver = gaussian, sigma = 1.0)
+Description: Rank the raw vector, shift the ranked Alpha vector, apply distribution (gaussian, cauchy, uniform). If driver is uniform, it simply subtract each Alpha value with the mean of all Alpha values in the Alpha vector
+Operator: rank(x, rate=2)
+Description: Ranks the input among all the instruments and returns an equally distributed number between 0.0 and 1.0. For precise sort, use the rate as 0
+Operator: scale(x, scale=1, longscale=1, shortscale=1)
+Description: Scales input to booksize. We can also scale the long positions and short positions to separate scales by mentioning additional parameters to the operator
+Operator: winsorize(x, std=4)
+Description: Winsorizes x to make sure that all values in x are between the lower and upper limits, which are specified as multiple of std.
+Operator: zscore(x)
+Description: Z-score is a numerical measurement that describes a value's relationship to the mean of a group of values. Z-score is measured in terms of standard deviations from the mean
+Operator: vec_avg(x)
+Description: Taking mean of the vector field x
+Operator: vec_sum(x)
+Description: Sum of vector field x
+Operator: bucket(rank(x), range="0, 1, 0.1" or buckets = "2,5,6,7,10")
+Description: Convert float values into indexes for user-specified buckets. Bucket is useful for creating group values, which can be passed to GROUP as input
+Operator: trade_when(x, y, z)
+Description: Used in order to change Alpha values only under a specified condition and to hold Alpha values in other cases. It also allows to close Alpha positions (assign NaN values) under a specified condition
+Operator: group_backfill(x, group, d, std = 4.0)
+Description: If a certain value for a certain date and instrument is NaN, from the set of same group instruments, calculate winsorized mean of all non-NaN values over last d days
+Operator: group_mean(x, weight, group)
+Description: All elements in group equals to the mean
+Operator: group_neutralize(x, group)
+Description: Neutralizes Alpha against groups. These groups can be subindustry, industry, sector, country or a constant
+Operator: group_rank(x, group)
+Description: Each elements in a group is assigned the corresponding rank in this group
+Operator: group_scale(x, group)
+Description: Normalizes the values in a group to be between 0 and 1. (x - groupmin) / (groupmax - groupmin)
+Operator: group_zscore(x, group)
+Description: Calculates group Z-score - numerical measurement that describes a value's relationship to the mean of a group of values. Z-score is measured in terms of standard deviations from the mean. zscore = (data - mean) / stddev of x for each instrument within its group.
+========================= 操作符结束 =======================================
+
+========================= 数据字段开始 =======================================注意: DataField: 后面的是数据字段, DataFieldDescription: 此字段后面的是数据字段对应的描述或使用说明, DataFieldDescription字段后面的内容是使用说明, 不是数据字段
+
+DataField: pcr_vol_720
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 720 days in the future.
+DataField: option_breakeven_90
+DataFieldDescription: Price at which a stock's options with expiration 90 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_270
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 270 days in the future.
+DataField: option_breakeven_360
+DataFieldDescription: Price at which a stock's options with expiration 360 days in the future break even based on its recent bid/ask mean.
+DataField: put_breakeven_150
+DataFieldDescription: Price at which a stock's put options with expiration 150 days in the future break even based on its recent bid/ask mean.
+DataField: option_breakeven_720
+DataFieldDescription: Price at which a stock's options with expiration 720 days in the future break even based on its recent bid/ask mean.
+DataField: call_breakeven_20
+DataFieldDescription: Price at which a stock's call options with expiration 20 days in the future break even based on its recent bid/ask mean.
+DataField: put_breakeven_180
+DataFieldDescription: Price at which a stock's put options with expiration 180 days in the future break even based on its recent bid/ask mean.
+DataField: call_breakeven_10
+DataFieldDescription: Price at which a stock's call options with expiration 10 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_1080
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 1080 days in the future.
+DataField: pcr_oi_30
+DataFieldDescription: Ratio of put open interest to call open interest on a stock's options with expiration 30 days in the future.
+DataField: option_breakeven_180
+DataFieldDescription: Price at which a stock's options with expiration 180 days in the future break even based on its recent bid/ask mean.
+DataField: option_breakeven_60
+DataFieldDescription: Price at which a stock's options with expiration 60 days in the future break even based on its recent bid/ask mean.
+DataField: forward_price_20
+DataFieldDescription: Forward price at 20 days derived from a synthetic long option with payoff similar to long stock + option dynamics. Combination of long ATM call and short ATM put.
+DataField: forward_price_360
+DataFieldDescription: Forward price at 360 days derived from a synthetic long option with payoff similar to long stock + option dynamics. Combination of long ATM call and short ATM put.
+DataField: forward_price_180
+DataFieldDescription: Forward price at 180 days derived from a synthetic long option with payoff similar to long stock + option dynamics. combination of long ATM call, and short ATM put.
+DataField: call_breakeven_1080
+DataFieldDescription: Price at which a stock's call options with expiration 1080 days in the future break even based on its recent bid/ask mean.
+DataField: put_breakeven_1080
+DataFieldDescription: Price at which a stock's put options with expiration 1080 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_180
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 180 days in the future.
+DataField: pcr_oi_120
+DataFieldDescription: Ratio of put open interest to call open interest on a stock's options with expiration 120 days in the future.
+DataField: call_breakeven_120
+DataFieldDescription: Price at which a stock's call options with expiration 120 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_90
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 90 days in the future.
+DataField: pcr_vol_20
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 20 days in the future.
+DataField: put_breakeven_60
+DataFieldDescription: Price at which a stock's put options with expiration 60 days in the future break even based on its recent bid/ask mean.
+DataField: put_breakeven_120
+DataFieldDescription: Price at which a stock's put options with expiration 120 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_oi_270
+DataFieldDescription: Ratio of put open interest to call open interest on a stock's options with expiration 270 days in the future.
+DataField: forward_price_1080
+DataFieldDescription: Forward price at 1080 days derived from a synthetic long option with payoff similar to long stock + option dynamics. Combination of long ATM call and short ATM put.
+DataField: option_breakeven_120
+DataFieldDescription: Price at which a stock's options with expiration 120 days in the future break even based on its recent bid/ask mean.
+DataField: pcr_vol_30
+DataFieldDescription: Ratio of put volume to call volume on a stock's options with expiration 30 days in the future.
+DataField: option_breakeven_150
+DataFieldDescription: Price at which a stock's options with expiration 150 days in the future break even based on its recent bid/ask mean.
+DataField: fnd6_prcc
+DataFieldDescription: Price Close - Annual
+DataField: fnd6_mkvalt
+DataFieldDescription: Market Value - Total
+DataField: fnd6_newqeventv110_txdiq
+DataFieldDescription: Income Taxes - Deferred
+DataField: fnd6_newa1v1300_cstk
+DataFieldDescription: Common/Ordinary Stock (Capital)
+DataField: fnd6_newqeventv110_tstkq
+DataFieldDescription: Treasury Stock - Total (All Capital)
+DataField: fnd6_cld3
+DataFieldDescription: Capitalized Leases - Due in 3rd Year
+DataField: fnd6_newqv1300_aqpl1q
+DataFieldDescription: Assets Level 1 (Quoted Prices)
+DataField: fnd6_newqv1300_txdiq
+DataFieldDescription: Income Taxes - Deferred
+DataField: fnd6_newa1v1300_epsfx
+DataFieldDescription: Earnings Per Share (Diluted) - Excluding Extraordinary Items
+DataField: fnd6_newqv1300_ivstq
+DataFieldDescription: Short-Term Investments - Total
+DataField: enterprise_value
+DataFieldDescription: Enterprise Value
+DataField: fnd6_state
+DataFieldDescription: integer for identifying the state of the company
+DataField: fnd6_spce
+DataFieldDescription: S&P Core Earnings
+DataField: fnd6_txdbclq
+DataFieldDescription: Current Deferred Tax Liability
+DataField: fnd6_newa1v1300_dvc
+DataFieldDescription: Dividends Common/Ordinary
+DataField: fnd6_eventv110_dd1q
+DataFieldDescription: Long Term Debt Due in 1 Year
+DataField: fnd6_xidos
+DataFieldDescription: Extraordinary Items and Discontinued Operations
+DataField: fnd6_newqv1300_spcedq
+DataFieldDescription: S&P Core Earnings EPS Diluted
+DataField: fnd6_newqeventv110_txtq
+DataFieldDescription: Income Taxes - Total
+DataField: fnd6_newqeventv110_xidoq
+DataFieldDescription: Extraordinary Items and Discontinued Operations
+DataField: operating_expense
+DataFieldDescription: Operating Expense - Total
+DataField: fnd6_newqeventv110_icaptq
+DataFieldDescription: Invested Capital - Total - Quarterly
+DataField: fnd6_newqeventv110_ibadj12
+DataFieldDescription: Income Before Extra Items - Adj for Common Stock Equivalents - 12MM
+DataField: fnd6_eventv110_xaccq
+DataFieldDescription: Accrued Expenses
+DataField: fnd6_pncdq
+DataFieldDescription: Core Pension Adjustment Diluted EPS Effect
+DataField: fnd6_eventv110_gdwlid12
+DataFieldDescription: Impairments Diluted EPS - 12mm
+DataField: fnd6_cld2
+DataFieldDescription: Capitalized Leases - Due in 2nd Year
+DataField: fnd6_oelim
+DataFieldDescription: Other Eliminations (Income)
+DataField: fnd6_esubs
+DataFieldDescription: Equity in Earnings
+DataField: fnd6_newqeventv110_xoptepsqp
+DataFieldDescription: Implied Option EPS Basic Preliminary
+DataField: scl12_alltype_buzzvec
+DataFieldDescription: sentiment volume
+DataField: scl12_alltype_sentvec
+DataFieldDescription: sentiment
+DataField: scl12_alltype_typevec
+DataFieldDescription: instrument type index
+DataField: scl12_buzz
+DataFieldDescription: relative sentiment volume
+DataField: scl12_buzz_fast_d1
+DataFieldDescription: relative sentiment volume
+DataField: scl12_buzzvec
+DataFieldDescription: sentiment volume
+DataField: scl12_sentiment
+DataFieldDescription: sentiment
+DataField: scl12_sentiment_fast_d1
+DataFieldDescription: sentiment
+DataField: scl12_sentvec
+DataFieldDescription: sentiment
+DataField: scl12_typevec
+DataFieldDescription: instrument type index
+DataField: snt_buzz
+DataFieldDescription: negative relative sentiment volume, fill nan with 0
+DataField: snt_buzz_bfl
+DataFieldDescription: negative relative sentiment volume, fill nan with 1
+DataField: snt_buzz_bfl_fast_d1
+DataFieldDescription: negative relative sentiment volume, fill nan with 1
+DataField: snt_buzz_fast_d1
+DataFieldDescription: negative relative sentiment volume, fill nan with 0
+DataField: snt_buzz_ret
+DataFieldDescription: negative return of relative sentiment volume
+DataField: snt_buzz_ret_fast_d1
+DataFieldDescription: negative return of relative sentiment volume
+DataField: snt_value
+DataFieldDescription: negative sentiment, fill nan with 0
+DataField: snt_value_fast_d1
+DataFieldDescription: negative sentiment, fill nan with 0
+DataField: analyst_revision_rank_derivative
+DataFieldDescription: Change in ranking for analyst revisions and momentum compared to previous period.
+DataField: cashflow_efficiency_rank_derivative
+DataFieldDescription: Change in ranking for cash flow generation and profitability compared to previous period.
+DataField: composite_factor_score_derivative
+DataFieldDescription: Change in overall composite factor score from the prior period.
+DataField: earnings_certainty_rank_derivative
+DataFieldDescription: Change in ranking for earnings sustainability and certainty compared to previous period.
+DataField: fscore_bfl_growth
+DataFieldDescription: The purpose of this metric is to qualify the expected MT growth potential of the stock.
+DataField: fscore_bfl_momentum
+DataFieldDescription: The purpose of this metric is to identify stocks which are currently undergoing either up or downward analyst revisions.
+DataField: fscore_bfl_profitability
+DataFieldDescription: The purpose of this metric is to rank stock based on their ability to generate cash flows.
+DataField: fscore_bfl_quality
+DataFieldDescription: The purpose of this metric is to measure both the sustainability and certainty of earnings.
+DataField: fscore_bfl_surface
+DataFieldDescription: The static score. An index between 0 & 100 is applied for each stock and each composite factor - The first ranking is a pentagon surface-based score. The larger the surface, the higher the rank.
+DataField: fscore_bfl_surface_accel
+DataFieldDescription: The derivative score. In a second step, we calculate the derivative of this score (ie: Is the surface of the pentagon increasing or decreasing from the previous month?).
+DataField: fscore_bfl_total
+DataFieldDescription: The final score M-Score is a weighted average of both the Pentagon surface score and the Pentagon acceleration score.
+DataField: fscore_bfl_value
+DataFieldDescription: The purpose of this metric is to see if the stock is under or overpriced given several well known valuation standards.
+DataField: fscore_growth
+DataFieldDescription: The purpose of this metric is to qualify the expected MT growth potential of the stock.
+DataField: fscore_momentum
+DataFieldDescription: The purpose of this metric is to identify stocks which are currently undergoing either up or downward analyst revisions.
+DataField: fscore_profitability
+DataFieldDescription: The purpose of this metric is to rank stock based on their ability to generate cash flows.
+DataField: fscore_quality
+DataFieldDescription: The purpose of this metric is to measure both the sustainability and certainty of earnings.
+DataField: fscore_surface
+DataFieldDescription: The static score. An index between 0 & 100 is applied for each stock and each composite factor - The first ranking is a pentagon surface-based score. The larger the surface, the higher the rank.
+DataField: fscore_surface_accel
+DataFieldDescription: The derivative score. In a second step, we calculate the derivative of this score (ie: Is the surface of the pentagon increasing or decreasing from the previous month?).
+DataField: fscore_total
+DataFieldDescription: The final score M-Score is a weighted average of both the Pentagon surface score and the Pentagon acceleration score.
+DataField: fscore_value
+DataFieldDescription: The purpose of this metric is to see if the stock is under or overpriced given several well known valuation standards.
+DataField: growth_potential_rank_derivative
+DataFieldDescription: Change in ranking for medium-term growth potential compared to previous period.
+DataField: multi_factor_acceleration_score_derivative
+DataFieldDescription: Change in the acceleration of multi-factor score compared to previous period.
+DataField: multi_factor_static_score_derivative
+DataFieldDescription: Change in static multi-factor score compared to previous period.
+DataField: relative_valuation_rank_derivative
+DataFieldDescription: Change in ranking for valuation metrics compared to previous period.
+DataField: snt_social_value
+DataFieldDescription: Z score of sentiment
+DataField: snt_social_volume
+DataFieldDescription: Normalized tweet volume
+DataField: beta_last_30_days_spy
+DataFieldDescription: Beta to SPY in 30 Days
+DataField: beta_last_360_days_spy
+DataFieldDescription: Beta to SPY in 360 Days
+DataField: beta_last_60_days_spy
+DataFieldDescription: Beta to SPY in 60 Days
+DataField: beta_last_90_days_spy
+DataFieldDescription: Beta to SPY in 90 Days
+DataField: correlation_last_30_days_spy
+DataFieldDescription: Correlation to SPY in 30 Days
+DataField: correlation_last_360_days_spy
+DataFieldDescription: Correlation to SPY in 360 Days
+DataField: correlation_last_60_days_spy
+DataFieldDescription: Correlation to SPY in 60 Days
+DataField: correlation_last_90_days_spy
+DataFieldDescription: Correlation to SPY in 90 Days
+DataField: systematic_risk_last_30_days
+DataFieldDescription: Systematic Risk Last 30 Days
+DataField: systematic_risk_last_360_days
+DataFieldDescription: Systematic Risk Last 360 Days
+DataField: systematic_risk_last_60_days
+DataFieldDescription: Systematic Risk Last 60 Days
+DataField: systematic_risk_last_90_days
+DataFieldDescription: Systematic Risk Last 90 Days
+DataField: unsystematic_risk_last_30_days
+DataFieldDescription: Unsystematic Risk Last 30 Days - Relative to SPY
+DataField: unsystematic_risk_last_360_days
+DataFieldDescription: Unsystematic Risk Last 360 Days - Relative to SPY
+DataField: unsystematic_risk_last_60_days
+DataFieldDescription: Unsystematic Risk Last 60 Days - Relative to SPY
+DataField: unsystematic_risk_last_90_days
+DataFieldDescription: Unsystematic Risk Last 90 Days - Relative to SPY
+DataField: anl4_afv4_div_number
+DataFieldDescription: Number of estimations for Dividend per share - annually
+DataField: sales_max_guidance_value
+DataFieldDescription: Maximum guidance value for annual sales
+DataField: anl4_fsguidanceqfv4_item
+DataFieldDescription: Financial item
+DataField: anl4_cff_number
+DataFieldDescription: Cash Flow From Financing - number of estimations
+DataField: anl4_fsguidanceafv4_maxguidance
+DataFieldDescription: Maximum guidance value
+DataField: anl4_tot_gw_ft
+DataFieldDescription: Total Goodwill - forecast type (revision/new/...)
+DataField: cash_flow_from_operations
+DataFieldDescription: Cash Flow from Operations - Value for the annual period
+DataField: anl4_dez1qfv4_est
+DataFieldDescription: Estimation value
+DataField: anl4_eaz2lltv110_estvalue
+DataFieldDescription: Estimation value
+DataField: anl4_cfo_mean
+DataFieldDescription: Cash Flow From Operations - mean of estimations
+DataField: anl4_fcfps_flag
+DataFieldDescription: Free cash flow per share - forecast type (revision/new/...)
+DataField: anl4_qfd1_az_wol_vid
+DataFieldDescription: Dividend per share - The lowest value among forecasts
+DataField: anl4_basicqfv4_minguidance
+DataFieldDescription: Min guidance value
+DataField: anl4_cuo1conafv110_item
+DataFieldDescription: Financial item
+DataField: anl4_qf_az_div_mean
+DataFieldDescription: Dividend per share - average of estimations
+DataField: min_stock_option_expense_guidance
+DataFieldDescription: Stock option expense - minimum guidance value
+DataField: anl4_bvps_flag
+DataFieldDescription: Book value per share - forecast type (revision/new/...)
+DataField: anl4_tbvps_number
+DataFieldDescription: Tangible Book Value per Share - number of estimations
+DataField: anl4_epsa_flag
+DataFieldDescription: Earnings per share adjusted by excluding extraordinary items and stock option expenses - forecast type (revision/new/...)
+DataField: min_reported_eps_guidance
+DataFieldDescription: Reported Earnings Per Share - Minimum guidance value for the annual period
+DataField: max_net_income_guidance
+DataFieldDescription: The maximum guidance value for net profit.
+DataField: anl4_netprofit_std
+DataFieldDescription: Net profit - standard deviation of estimations
+DataField: anl4_qf_az_dts_spe
+DataFieldDescription: Earnings per share - std of estimations
+DataField: anl4_adxqfv110_pu
+DataFieldDescription: The number of upper estimations
+DataField: anl4_netprofit_value
+DataFieldDescription: Net profit- announced financial value
+DataField: cashflow_per_share_max_guidance_quarterly
+DataFieldDescription: The maximum guidance value for Cash Flow Per Share.
+DataField: anl4_basicdetailqfv110_person
+DataFieldDescription: Broker Id
+DataField: anl4_fsguidanceafv4_item
+DataFieldDescription: Financial item
+DataField: min_shares_outstanding_guidance
+DataFieldDescription: Minimum guidance value for Shares
+DataField: anl4_qfv4_div_low
+DataFieldDescription: Dividend per share - The lowest estimation
+DataField: pv13_hierarchy23_sector
+DataFieldDescription: grouping fields
+DataField: pv13_ustomergraphrank_auth_rank
+DataFieldDescription: the HITS authority score of customers
+DataField: pv13_rha2_min2_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_rha2_min5_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min5_1000_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_new_5l_scibr
+DataFieldDescription: grouping fields
+DataField: pv13_revere_level
+DataFieldDescription: Level of the sector within the hierarchy
+DataField: pv13_hierarchy_min10_sector_3000_sector
+DataFieldDescription: grouping fields
+DataField: pv13_revere_index_cap
+DataFieldDescription: Company market capitalization
+DataField: pv13_rha2_min2_1000_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_rha2_min2_sector
+DataFieldDescription: grouping fields
+DataField: pv13_di_5l
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchys32_sector
+DataFieldDescription: grouping fields
+DataField: pv13_ustomergraphrank_page_rank
+DataFieldDescription: the PageRank of customers
+DataField: pv13_h_min24_500_sector
+DataFieldDescription: Grouping fields for top 500
+DataField: pv13_hierarchy_min30_3000_mapped_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_rha2_min5_3000_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min25_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min10_sector_3000_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min51_f4_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min5_f3g2_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min20_3k_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min10_industry_3000_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min52_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_r2_min20_1000_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy2_513_sector
+DataFieldDescription: grouping fields
+DataField: pv13_h2_min2_1k_sector
+DataFieldDescription: Grouping fields for top 1000
+DataField: pv13_r2_min10_3000_sector
+DataFieldDescription: grouping fields
+DataField: pv13_hierarchy_min30_3000_513_sector
+DataFieldDescription: grouping fields
+DataField: implied_volatility_mean_skew_60
+DataFieldDescription: At-the-money option-implied volatility mean skew for 60 days
+DataField: parkinson_volatility_180
+DataFieldDescription: Parkinson model's historical volatility over 180 days
+DataField: parkinson_volatility_30
+DataFieldDescription: Parkinson model's historical volatility over 30 days
+DataField: parkinson_volatility_90
+DataFieldDescription: Parkinson model's historical volatility over 90 days
+DataField: implied_volatility_mean_150
+DataFieldDescription: At-the-money option-implied volatility mean for 150 days
+DataField: parkinson_volatility_10
+DataFieldDescription: Parkinson model's historical volatility over 2 weeks
+DataField: implied_volatility_put_60
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 60 days
+DataField: implied_volatility_call_60
+DataFieldDescription: At-the-money option-implied volatility for call Option for 60 days
+DataField: implied_volatility_put_30
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 30 days
+DataField: parkinson_volatility_120
+DataFieldDescription: Parkinson model's historical volatility over 120 days
+DataField: implied_volatility_mean_skew_270
+DataFieldDescription: At-the-money option-implied volatility mean skew for 270 days
+DataField: implied_volatility_put_90
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 90 days
+DataField: implied_volatility_call_90
+DataFieldDescription: At-the-money option-implied volatility for call Option for 90 days
+DataField: implied_volatility_put_1080
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 3 years
+DataField: implied_volatility_mean_skew_30
+DataFieldDescription: At-the-money option-implied volatility mean skew for 30 days
+DataField: parkinson_volatility_20
+DataFieldDescription: Parkinson model's historical volatility over 20 days
+DataField: implied_volatility_mean_60
+DataFieldDescription: At-the-money option-implied volatility mean for 60 days
+DataField: implied_volatility_call_10
+DataFieldDescription: At-the-money option-implied volatility for call Option for 10 days
+DataField: historical_volatility_120
+DataFieldDescription: Close-to-close Historical volatility over 120 days
+DataField: implied_volatility_mean_skew_10
+DataFieldDescription: At-the-money option-implied volatility mean skew for 10 days
+DataField: implied_volatility_put_360
+DataFieldDescription: At-the-money option-implied volatility for Put Option for 360 days
+DataField: parkinson_volatility_150
+DataFieldDescription: Parkinson model's historical volatility over 150 days
+DataField: historical_volatility_150
+DataFieldDescription: Close-to-close Historical volatility over 150 days
+DataField: historical_volatility_20
+DataFieldDescription: Close-to-close Historical volatility over 20 days
+DataField: implied_volatility_mean_20
+DataFieldDescription: At-the-money option-implied volatility mean for 20 days
+DataField: implied_volatility_call_20
+DataFieldDescription: At-the-money option-implied volatility for call Option for 20 days
+DataField: historical_volatility_30
+DataFieldDescription: Close-to-close Historical volatility over 30 days
+DataField: implied_volatility_call_720
+DataFieldDescription: At-the-money option-implied volatility for call Option for 720 days
+DataField: implied_volatility_mean_720
+DataFieldDescription: At-the-money option-implied volatility mean for 720 days
+DataField: implied_volatility_call_360
+DataFieldDescription: At-the-money option-implied volatility for call Option for 360 days
+DataField: news_indx_perf
+DataFieldDescription: ((EODClose - TONLast) / TONLast) - ((SPYClose - SPYLast) / SPYLast)
+DataField: news_mins_7_5_pct_dn
+DataFieldDescription: Number of minutes that elapsed before price went down 7.5 percentage points
+DataField: nws12_prez_result_vs_index
+DataFieldDescription: ((EODClose - TONLast) / TONLast) - ((SPYClose - SPYLast) / SPYLast)
+DataField: nws12_prez_2s
+DataFieldDescription: Number of minutes that elapsed before price went down 2 percentage points
+DataField: nws12_mainz_dayopen
+DataFieldDescription: Price at the session open
+DataField: nws12_prez_5l
+DataFieldDescription: Number of minutes that elapsed before price went up 5 percentage points
+DataField: nws12_mainz_02l
+DataFieldDescription: Number of minutes that elapsed before price went up 20 percentage points
+DataField: news_short_interest
+DataFieldDescription: Total number of shares sold short divided by total number of shares outstanding
+DataField: nws12_mainz_maxdnamt
+DataFieldDescription: The price at the time of the news minus the after the news low
+DataField: news_close_vol
+DataFieldDescription: Main close volume
+DataField: nws12_prez_maxup
+DataFieldDescription: Percent change from the price at the time of the news to the after the news high
+DataField: nws12_afterhsz_2l
+DataFieldDescription: Number of minutes that elapsed before price went up 2 percentage points
+DataField: nws12_afterhsz_range
+DataFieldDescription: Session High Price - Session Low Price) / Session Low Price.
+DataField: nws12_mainz_1s
+DataFieldDescription: Number of minutes that elapsed before price went down 1 percentage point
+DataField: nws12_prez_spylast
+DataFieldDescription: Last Price of the SPY at the time of the news
+DataField: nws12_prez_result2
+DataFieldDescription: Percent change between the price at the time of the news release to the price at the close of the session
+DataField: nws12_mainz_mktcap
+DataFieldDescription: Reported market capitalization for the calendar day of the session
+DataField: news_max_up_ret
+DataFieldDescription: Percent change from the price at the time of the news to the after the news high
+DataField: nws12_afterhsz_5_min
+DataFieldDescription: The percent change in price in the first 5 minutes following the news release
+DataField: nws12_mainz_01s
+DataFieldDescription: Number of minutes that elapsed before price went down 10 percentage points
+DataField: nws12_mainz_newrecord
+DataFieldDescription: Tracks whether the news is first instance or a duplicate
+DataField: nws12_mainz_provider
+DataFieldDescription: index of name of the news provider
+DataField: news_pct_1min
+DataFieldDescription: The percent change in price in the first minute following the news release
+DataField: nws12_mainz_1_minute
+DataFieldDescription: The percent change in price in the first minute following the news release
+DataField: nws12_mainz_range
+DataFieldDescription: Session High Price - Session Low Price) / Session Low Price.
+DataField: nws12_prez_prev_vol
+DataFieldDescription: Previous day's session volume
+DataField: nws12_mainz_5_min
+DataFieldDescription: The percent change in price in the first 5 minutes following the news release
+DataField: nws12_mainz_5l
+DataFieldDescription: Number of minutes that elapsed before price went up 5 percentage points
+DataField: nws12_mainz_4p
+DataFieldDescription: The minimum of L or S above for 4 minute bucket
+DataField: nws12_mainz_57l
+DataFieldDescription: Number of minutes that elapsed before price went up 7.5 percentage points
+DataField: top1000
+DataFieldDescription: 20140630
+DataField: top200
+DataFieldDescription: 20140630
+DataField: top3000
+DataFieldDescription: 20140630
+DataField: top500
+DataFieldDescription: 20140630
+DataField: topsp500
+DataFieldDescription: 20140630
+DataField: rp_ess_technical
+DataFieldDescription: Event sentiment score based on technical analysis
+DataField: rp_nip_assets
+DataFieldDescription: News impact projection of assets news
+DataField: rp_ess_price
+DataFieldDescription: Event sentiment score of stock price news
+DataField: nws18_event_similarity_days
+DataFieldDescription: Days since a similar event was detected
+DataField: nws18_ssc
+DataFieldDescription: Sentiment of the news calculated using multiple techniques
+DataField: nws18_acb
+DataFieldDescription: News sentiment specializing in corporate action announcements
+DataField: rp_ess_product
+DataFieldDescription: Event sentiment score of product and service-related news
+DataField: rp_nip_labor
+DataFieldDescription: News impact projection of labor issues news
+DataField: nws18_bam
+DataFieldDescription: News sentiment specializing in mergers and acquisitions
+DataField: rp_css_business
+DataFieldDescription: Composite sentiment score of business-related news
+DataField: rp_css_partner
+DataFieldDescription: Composite sentiment score of partnership news
+DataField: rp_css_labor
+DataFieldDescription: Composite sentiment score of labor issues news
+DataField: rp_css_assets
+DataFieldDescription: Composite sentiment score of assets news
+DataField: nws18_ghc_lna
+DataFieldDescription: Change in analyst recommendation
+DataField: rp_nip_society
+DataFieldDescription: News impact projection of society-related news
+DataField: rp_nip_marketing
+DataFieldDescription: News impact projection of marketing news
+DataField: rp_ess_credit_ratings
+DataFieldDescription: Event sentiment score of credit ratings news
+DataField: rp_css_insider
+DataFieldDescription: Composite sentiment score of insider trading news
+DataField: rp_nip_partner
+DataFieldDescription: News impact projection of partnership news
+DataField: rp_css_ratings
+DataFieldDescription: Composite sentiment score of analyst ratings-related news
+DataField: rp_css_earnings
+DataFieldDescription: Composite sentiment score of earnings news
+DataField: rp_nip_ratings
+DataFieldDescription: News impact projection of analyst ratings-related news
+DataField: rp_nip_credit_ratings
+DataFieldDescription: News impact projection of credit ratings news
+DataField: rp_ess_dividends
+DataFieldDescription: Event sentiment score of dividends news
+DataField: rp_css_price
+DataFieldDescription: Composite sentiment score of stock price news
+DataField: rp_ess_partner
+DataFieldDescription: Event sentiment score of partnership news
+DataField: rp_css_product
+DataFieldDescription: Composite sentiment score of product and service-related news
+DataField: rp_ess_equity
+DataFieldDescription: Event sentiment score of equity action news
+DataField: rp_nip_technical
+DataFieldDescription: News impact projection based on technical analysis
+DataField: rp_nip_inverstor
+DataFieldDescription: News impact projection of investor relations news
+DataField: fn_prepaid_expense_q
+DataFieldDescription: Carrying amount for an unclassified balance sheet date of expenditures made in advance of when the economic benefit of the cost will be realized, and which will be expensed in future periods with the passage of time or when a triggering event occurs. For a classified balance sheet, represents the noncurrent portion of prepaid expenses (the current portion has a separate concept).
+DataField: fn_oth_income_loss_net_of_tax_a
+DataFieldDescription: Amount after tax and reclassification adjustments of other comprehensive income (loss).
+DataField: fnd2_dbplanepdfbnfp5ytherea
+DataFieldDescription: Amount of benefits from a defined benefit plan expected to be paid in the 5 fiscal years after the 5th fiscal year following the latest fiscal year. Excludes interim and annual periods when interim periods are reported on a rolling approach, from latest balance sheet date.
+DataField: fn_income_from_equity_investments_a
+DataFieldDescription: Income From Equity Method Investments
+DataField: fn_accrued_liab_curr_q
+DataFieldDescription: Carrying value as of the balance sheet date of obligations incurred and payable, pertaining to costs that are statutory in nature, are incurred on contractual obligations, or accumulate over time and for which invoices have not yet been received or will not be rendered.
+DataField: fnd2_asdm
+DataFieldDescription: Assets, Domestic
+DataField: fnd2_a_eplsbvdcpcstnrgprg
+DataFieldDescription: The weighted average period over which unrecognized compensation is expected to be recognized for equity-based compensation plans, using a decimal to express in number of years.
+DataField: fn_assets_fair_val_l3_q
+DataFieldDescription: Asset Fair Value, Recurring, Level 3
+DataField: fn_op_lease_min_pay_due_in_5y_a
+DataFieldDescription: Amount of required minimum rental payments for operating leases having an initial or remaining non-cancelable lease term in excess of 1 year due in the 5th fiscal year following the latest fiscal year. Excludes interim and annual periods when interim periods are reported on a rolling approach, from latest balance sheet date.
+DataField: fnd2_unremittedfrer
+DataFieldDescription: Unremitted Foreign Earnings
+DataField: fnd2_propplteqmuflmamfrt
+DataFieldDescription: PPE, Furniture, Useful Life, Maximum
+DataField: fn_def_income_tax_expense_q
+DataFieldDescription: Income Tax Expense, Deferred
+DataField: fnd2_dfdfeditxexp
+DataFieldDescription: Income Tax Expense, Deferred - Federal
+DataField: fn_assets_fair_val_l1_q
+DataFieldDescription: Asset Fair Value, Recurring, Level 1
+DataField: fnd2_dbplanamtsrginblsh
+DataFieldDescription: The aggregate net amount recognized in the balance sheet associated with the defined benefit plan(s). Will normally be the same as the Defined Benefit Plan, Funded Status of Plan, Total.
+DataField: fn_treasury_stock_shares_a
+DataFieldDescription: Number of common and preferred shares that were previously issued and that were repurchased by the issuing entity and held in treasury on the financial statement date. This stock has no voting rights and receives no dividends.
+DataField: fn_comp_options_exercises_weighted_avg_a
+DataFieldDescription: Share-Based Compensation, Options Assumed, Weighted Average Exercise Price
+DataField: fn_proceeds_from_stock_options_exercised_q
+DataFieldDescription: The cash inflow associated with the amount received from holders exercising their stock options. This item inherently excludes any excess tax benefit, which the entity may have realized and reported separately.
+DataField: fn_assets_fair_val_l2_a
+DataFieldDescription: Asset Fair Value, Recurring, Level 2
+DataField: fn_interest_payable_a
+DataFieldDescription: Carrying value as of the balance sheet date of [accrued] interest payable on all forms of debt, including trade payables, that has been incurred and is unpaid. For classified balance sheets, used to reflect the current portion of the liabilities (due within 1 year or within the normal operating cycle if longer); for unclassified balance sheets, used to reflect the total liabilities (regardless of due date).
+DataField: fn_repurchased_shares_value_a
+DataFieldDescription: Shares repurchased and either retired or put into treasury stock, likely as part of a share buyback plan.
+DataField: fn_op_lease_rent_exp_a
+DataFieldDescription: Rental expense for the reporting period incurred under operating leases, including minimum and any contingent rent expense, net of related sublease income.
+DataField: fn_debt_instrument_interest_rate_stated_percentage_a
+DataFieldDescription: Stated percentage of interest rate on debt
+DataField: fn_new_shares_issued_a
+DataFieldDescription: Number of new stock issued during the period.
+DataField: fnd2_propplteqmuflmeqmt
+DataFieldDescription: PPE, Equipment, Useful Life, Minimum
+DataField: fnd2_propplteqmuflmblgland
+DataFieldDescription: PPE, Buildings & land, Useful Life, Minimum
+DataField: fn_goodwill_acquired_during_period_a
+DataFieldDescription: Amount of increase in asset representing future economic benefits arising from other assets acquired in a business combination that are not individually identified and separately recognized resulting from a business combination.
+DataField: fn_finite_lived_intangible_assets_net_a
+DataFieldDescription: Finite Lived Intangible Assets, Net
+DataField: fn_business_combination_purchase_price_q
+DataFieldDescription: Business Combination, Purchase Price
+DataField: fn_liab_fair_val_l2_q
+DataFieldDescription: Liabilities Fair Value, Recurring, Level 2
+DataField: adv20
+DataFieldDescription: Average daily volume in past 20 days
+DataField: cap
+DataFieldDescription: Daily market capitalization (in millions)
+DataField: close
+DataFieldDescription: Daily close price
+DataField: country
+DataFieldDescription: Country grouping
+DataField: currency
+DataFieldDescription: Currency
+DataField: cusip
+DataFieldDescription: CUSIP Value
+DataField: dividend
+DataFieldDescription: Dividend
+DataField: exchange
+DataFieldDescription: Exchange grouping
+DataField: high
+DataFieldDescription: Daily high price
+DataField: industry
+DataFieldDescription: Industry grouping
+DataField: isin
+DataFieldDescription: ISIN Value
+DataField: low
+DataFieldDescription: Daily low price
+DataField: market
+DataFieldDescription: Market grouping
+DataField: open
+DataFieldDescription: Daily open price
+DataField: returns
+DataFieldDescription: Daily returns
+DataField: sector
+DataFieldDescription: Sector grouping
+DataField: sedol
+DataFieldDescription: Sedol
+DataField: sharesout
+DataFieldDescription: Daily outstanding shares (in millions)
+DataField: split
+DataFieldDescription: Stock split ratio
+DataField: subindustry
+DataFieldDescription: Subindustry grouping
+DataField: ticker
+DataFieldDescription: Ticker
+DataField: volume
+DataFieldDescription: Daily volume
+DataField: vwap
+DataFieldDescription: Daily volume weighted average price
+========================= 数据字段结束 =======================================
+

prepare_prompt/alpha_prompt.txt

@@ -1,36 +1,40 @@
 任务指令
-你是一个WorldQuant WebSim因子工程师。你的任务是生成 100 个用于行业轮动策略的复合型Alpha因子表达式。
+你是一个WorldQuant WebSim因子工程师。你的任务是生成 20 个用于行业轮动策略的复合型Alpha因子表达式。
 核心规则
 设计维度框架
-维度1：时间序列动量（TM）
-目标：识别价格趋势的强度、速度和持续性
-可用的具体构建方法：
-1. 简单动量：ts_delta(close, d) [d=5,10,20,30,60]
-2. 趋势斜率：ts_regression(close, ts_step(1), d, 0, 1) [rettype=1获取斜率]
-3. 动量加速度：ts_delta(ts_delta(close, d1), d2) [避免嵌套ts_regression]
-4. 平滑动量：ts_mean(returns, d) [returns=ts_delta(close,1)]
-5. 动量衰减：ts_decay_linear(returns, d)
-6. 价量关系：ts_corr(ts_delta(close,5), ts_delta(volume,5), d)
-建议组合：使用不同d参数创建短期/中期/长期动量
-
-维度2：横截面领导力（CL）
-目标：识别行业内的龙头股和相对强度
-具体构建方法：
-1. 龙头股筛选：if_else(rank(volume) > 0.7, 龙头值, 其他值) [使用volume代替market_cap]
-2. 龙头组合：group_mean(x, 1, bucket(rank(volume), range="0,3,0.4")) [使用volume排序]
-3. 行业内离散度：ts_std_dev(group_rank(returns, industry), 20)
-4. 相对排名稳定性：ts_mean(rank(returns), d)
-
-维度3：市场状态适应性（MS）
-目标：根据波动率、趋势状态调整参数
-具体构建方法：
-1. 波动率调整：ts_delta(close,5) / ts_std_dev(returns,20)
-2. 状态条件选择：if_else(ts_rank(volatility,30) > 0.7, 短期动量, 长期动量)
-3. 参数动态化：if_else(ts_std_dev(returns,20) > 阈值, 5, 20) [作为d参数]
-4. 趋势状态识别：ts_rank(ts_mean(returns,20), 60) > 0.5
-
-基本结构：
-复合因子 = 维度A组件 [运算符] 维度B组件 [条件调整]
+视角一：市场摩擦的横截面成像 (Cross-sectional Imaging of Market Frictions)
+核心洞见：传统因子假设市场无摩擦。真正的Alpha可能藏于“摩擦本身”——即指令流（order flow）转化为价格运动时，在不同股票上表现出的效率差异图谱。我们不为消除摩擦建模，而是主动测绘它。
+
+研究提示词 (用于指导因子构建)：
+
+指令流冲击的“消化速率”：测量单位异常交易量（可定义为其z-score）所引发的瞬时价格冲击（如未来1-5分钟收益率），以及该冲击在随后较短时间（如30分钟、1小时）的衰减半衰期。寻找“冲击大但衰减慢”（高摩擦、低流动性）与“冲击小但衰减快”（低摩擦、高流动性）的股票，并研究其横截面收益预测能力。
+
+买卖失衡的“路径依赖”：考察买单流与卖单流（可用正负交易额近似）的净额在时间序列上的自相关性模式。是呈现“均值回归”（失衡迅速被套利消化）还是“趋势持续”（失衡自我强化）？这种模式在不同波动率、不同市值股票中如何变化？能否构建一个度量“订单流趋势持续性”的指标？
+
+价格发现的“领地性”：将价格变化分解为“由自身交易驱动”的部分和“由同行业龙头/指数驱动”的部分（可通过日内收益率与龙头/指数收益率的滚动协方差分解）。计算“自身解释比例”。该比例高的股票，其价格发现是“内生性”的；比例低的则是“外生性”或“跟随性”的。这两类股票在不同市场环境（牛市/熊市/震荡市）下的表现有何系统性差异？
+
+视角二：投资者注意力的生态学建模 (Ecology of Investor Attention)
+核心洞见：注意力是金融市场最稀缺的资源。市场不是信息的聚合器，而是注意力资源的分配系统。因子应捕捉注意力的“聚集-分散”、“转移-停滞”动态，而非信息本身。
+
+研究提示词 (用于指导因子构建)：
+
+注意力“聚焦度”与“涣散度”：用交易量在时间序列上的分布熵来衡量。例如，过去20个交易日内，交易量分布的基尼系数或赫芬达尔指数。高集中度（某几天放巨量）意味着注意力短期爆发；低集中度（量能均匀）意味着持续平淡的关注。研究这两种状态转换前后，股票的收益特征。
+
+行业内注意力的“捕食-被捕食”关系：定义“注意力领导者”（如当日涨幅前3的股票）和“注意力追随者”（同行业其他股票）。计算领导股出现后，追随者的成交量与价格在随后N个时间段内的响应速度和强度。这种响应的不对称性（有些股票迅速响应，有些滞后）能否预测未来的相对强弱？
+
+“注意力惯性”与“注意力反转”：度量一个股票在失去短期催化剂（如公告、事件）后，其异常成交量（或搜索量、社交媒体活跃度）回落至基线水平的速度。回落慢的股票具有“注意力惯性”，可能存在定价偏差的持续；回落快的股票则“注意力反转”迅速。构建一个“注意力衰减时间”因子。
+
+视角三：价格运动的“形态语法”与“叙事连贯性” (Morphological Syntax & Narrative Coherence of Price Movements)
+核心洞见：价格运动不仅是有方向的，更是有“形状”和“语法”的。如同语言，一段价格走势是否存在合“语法”的叙事结构（如“筑底-突破-回踩-主升”），还是杂乱无章的“噪音词汇”？市场参与者会潜意识地识别并交易这些“连贯叙事”。
+
+研究提示词 (用于指导因子构建)：
+
+K线序列的“可压缩性”：使用一种简化的算法（如趋势线拟合后的残差大小，或一定容忍度下的分段线性近似所需节点数），来衡量过去一段价格序列的信息复杂度。低复杂度（高可压缩性）意味着价格运动清晰、有规律；高复杂度则意味着混乱。研究“从混乱转向清晰”或“从清晰转向混乱”的拐点。
+
+关键价位互动的“叙事逻辑”：识别近期的显著高点、低点、缺口、密集成交区作为“关键叙事节点”。分析当前价格在接近这些节点时的行为：是“尊重”（反弹/受阻）还是“无视”（直接穿越）？一系列对历史节点的“尊重”行为构成了一个连贯的“支撑阻力叙事”。能否量化这种叙事逻辑的强度？
+
+多尺度运动的“相位同步”：选取短期（如5日）、中期（20日）、长期（60日）的价格滤波序列（如移动平均线）。计算它们之间方向变化的领先滞后关系与同步性（可类似相位角分析）。例如，短期均线是否总是率先转向并引导中长期？还是三者经常背离？“多周期共振向上/向下”这种高度同步的状态，其形成过程和瓦解过程有何预测意义？
+
 
 === 关键语法规则（必须遵守） ===
 1. 数据字段规范：