QTR9090400 holdout_sensitivity_1.25_2_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop1.25ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return0.86
drawdown12.45
win rate47.29
sample332
mean trade pct0.01
average win pct1.69
average loss pct1.49
CI low pct-0.32
CI high pct0.35
quarter Kelly0
cost bps side6
calendar windows83
QTR9090400 holdout_sensitivity_1_2_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop1ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-21.58
drawdown25.81
win rate37.65
sample332
mean trade pct-0.29
average win pct1.61
average loss pct1.44
CI low pct-0.6
CI high pct0.03
quarter Kelly0
cost bps side16
calendar windows83
QTR9090400 holdout_sensitivity_1_2_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop1ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-7.42
drawdown14.74
win rate40.36
sample332
mean trade pct-0.09
average win pct1.7
average loss pct1.3
CI low pct-0.4
CI high pct0.23
quarter Kelly0
cost bps side6
calendar windows83
QTR9090400 holdout_sensitivity_0.75_2_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop0.75ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-23.34
drawdown26.49
win rate31.63
sample332
mean trade pct-0.32
average win pct1.66
average loss pct1.23
CI low pct-0.59
CI high pct-0.02
quarter Kelly0
cost bps side16
calendar windows83
QTR9090400 holdout_sensitivity_0.75_2_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop0.75ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-9.5
drawdown14.71
win rate33.73
sample332
mean trade pct-0.12
average win pct1.75
average loss pct1.07
CI low pct-0.39
CI high pct0.18
quarter Kelly0
cost bps side6
calendar windows83
QTR9090400 holdout_sensitivity_0.75_1.5_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop0.75ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-25.1
drawdown27.26
win rate34.34
sample332
mean trade pct-0.34
average win pct1.37
average loss pct1.24
CI low pct-0.59
CI high pct-0.09
quarter Kelly0
cost bps side16
calendar windows83
QTR9090400 holdout_sensitivity_0.75_1.5_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop0.75ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-11.57
drawdown15.28
win rate36.45
sample332
mean trade pct-0.14
average win pct1.49
average loss pct1.08
CI low pct-0.39
CI high pct0.11
quarter Kelly0
cost bps side6
calendar windows83
QTR9090400 holdout_trail_1_1.5_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule trail: last-session open; lagged20sessionmean(high-low) stop1ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; trail1ADRactivatedafter+1ADR and updatednextday, NOT15minuteimplementation. Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-23.53
drawdown26.31
win rate39.46
sample332
mean trade pct-0.32
average win pct1.33
average loss pct1.39
CI low pct-0.59
CI high pct-0.03
quarter Kelly0
cost bps side16
calendar windows83
QTR9090400 holdout_trail_1_1.5_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule trail: last-session open; lagged20sessionmean(high-low) stop1ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; trail1ADRactivatedafter+1ADR and updatednextday, NOT15minuteimplementation. Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-9.72
drawdown15.41
win rate42.47
sample332
mean trade pct-0.12
average win pct1.44
average loss pct1.26
CI low pct-0.39
CI high pct0.17
quarter Kelly0
cost bps side6
calendar windows83
QTR9090400 holdout_managed_1_1.5_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule managed: last-session open; lagged20sessionmean(high-low) stop1ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-24.07
drawdown27.08
win rate40.36
sample332
mean trade pct-0.33
average win pct1.34
average loss pct1.46
CI low pct-0.61
CI high pct-0.04
quarter Kelly0
cost bps side16
calendar windows83
QTR9090400 holdout_managed_1_1.5_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule managed: last-session open; lagged20sessionmean(high-low) stop1ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-10.36
drawdown16.54
win rate43.07
sample332
mean trade pct-0.13
average win pct1.45
average loss pct1.32
CI low pct-0.41
CI high pct0.16
quarter Kelly0
cost bps side6
calendar windows83
QTR9090400 holdout_open_0_0_16
BacktestsinconclusiveQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule open: last-session open to nextmonththirdclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return5.18
drawdown19.02
win rate51.81
sample332
mean trade pct0.08
average win pct1.85
average loss pct1.83
CI low pct-0.42
CI high pct0.56
quarter Kelly0.01
cost bps side16
calendar windows83
QTR9090400 holdout_open_0_0_6
BacktestsinconclusiveQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule open: last-session open to nextmonththirdclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return24.17
drawdown15.94
win rate58.13
sample332
mean trade pct0.28
average win pct1.84
average loss pct1.9
CI low pct-0.22
CI high pct0.76
quarter Kelly0.04
cost bps side6
calendar windows83
QTR9090400 holdout_parent_0_0_16
BacktestsinconclusiveQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule parent: penultimate-session close to nextmonththirdclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-9.22
drawdown22.42
win rate51.2
sample332
mean trade pct-0.1
average win pct1.66
average loss pct1.95
CI low pct-0.57
CI high pct0.36
quarter Kelly0
cost bps side16
calendar windows83
QTR9090400 holdout_parent_0_0_6
BacktestsinconclusiveQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. holdout test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule parent: penultimate-session close to nextmonththirdclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return7.17
drawdown14.64
win rate56.33
sample332
mean trade pct0.1
average win pct1.7
average loss pct1.97
CI low pct-0.37
CI high pct0.56
quarter Kelly0.01
cost bps side6
calendar windows83
QTR9090400 train_sensitivity_1.25_2_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop1.25ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-16.95
drawdown19.75
win rate37.5
sample136
mean trade pct-0.54
average win pct1
average loss pct1.46
CI low pct-0.91
CI high pct-0.16
quarter Kelly0
cost bps side16
calendar windows34
QTR9090400 train_sensitivity_1.25_2_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop1.25ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-11.11
drawdown14.53
win rate38.97
sample136
mean trade pct-0.34
average win pct1.16
average loss pct1.29
CI low pct-0.71
CI high pct0.04
quarter Kelly0
cost bps side6
calendar windows34
QTR9090400 train_sensitivity_1_2_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop1ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-17.96
drawdown21.16
win rate30.88
sample136
mean trade pct-0.57
average win pct1.05
average loss pct1.3
CI low pct-0.91
CI high pct-0.22
quarter Kelly0
cost bps side16
calendar windows34
QTR9090400 train_sensitivity_1_2_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop1ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-12.18
drawdown16.03
win rate31.62
sample136
mean trade pct-0.37
average win pct1.23
average loss pct1.11
CI low pct-0.71
CI high pct-0.02
quarter Kelly0
cost bps side6
calendar windows34
QTR9090400 train_sensitivity_0.75_2_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop0.75ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-16.46
drawdown19.92
win rate25.74
sample136
mean trade pct-0.52
average win pct1.12
average loss pct1.09
CI low pct-0.8
CI high pct-0.21
quarter Kelly0
cost bps side16
calendar windows34
QTR9090400 train_sensitivity_0.75_2_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop0.75ADR/target2ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-10.59
drawdown14.72
win rate26.47
sample136
mean trade pct-0.32
average win pct1.29
average loss pct0.9
CI low pct-0.6
CI high pct-0.01
quarter Kelly0
cost bps side6
calendar windows34
QTR9090400 train_sensitivity_0.75_1.5_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop0.75ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-15.8
drawdown19.64
win rate30.15
sample136
mean trade pct-0.5
average win pct0.92
average loss pct1.11
CI low pct-0.76
CI high pct-0.21
quarter Kelly0
cost bps side16
calendar windows34
QTR9090400 train_sensitivity_0.75_1.5_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule sensitivity: last-session open; lagged20sessionmean(high-low) stop0.75ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-9.88
drawdown14.41
win rate30.15
sample136
mean trade pct-0.3
average win pct1.12
average loss pct0.91
CI low pct-0.57
CI high pct-0.01
quarter Kelly0
cost bps side6
calendar windows34
QTR9090400 train_trail_1_1.5_16
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage15bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule trail: last-session open; lagged20sessionmean(high-low) stop1ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; trail1ADRactivatedafter+1ADR and updatednextday, NOT15minuteimplementation. Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-15.36
drawdown19.56
win rate30.88
sample136
mean trade pct-0.49
average win pct0.93
average loss pct1.12
CI low pct-0.8
CI high pct-0.17
quarter Kelly0
cost bps side16
calendar windows34
QTR9090400 train_trail_1_1.5_6
BacktestsrejectQuant replicationSPY, IWM, EFA, EEM2026-09-09 04:09
Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. train test, onlyentries AND exits within partition; holdout2019–25 has83 completecalendarwindows/332correlated ETFtrades, boundaryspanningtradesexcluded. Fourindependent initiallyequal250USD cashsleeves, no leverage, no periodiccross-sleeverebalance, idlecash0. Vectorbt from_orders explicitlongonly, fee1bp plus slippage5bp/side; rawcash-return reconcileswithVBT tradeP&L accounting for entryfee denominator. Rule trail: last-session open; lagged20sessionmean(high-low) stop1ADR/target1.5ADR; otherwisethirdsessionclose. Managed dailyOHLC stop-first ambiguity, adversegap fillsopen, targetgapcappedtarget; trail1ADRactivatedafter+1ADR and updatednextday, NOT15minuteimplementation. Dailybar touchfills preliminary; no claimed15minexecution, nor whole-lot/currentcap replication. Bootstrap10000 completecalendarwindowmeans acrossfourETFs, not332independenttrades; CI and per-symbol statistics saved observationmatchingtitle. Verdict rejects liveadoption when uncertainty/cost/executiongatesfail; positivepointestimatealone isnotadoption. Source https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://runway.adityaperswal.workers.dev/data/stock-bars .
return-9.41
drawdown14.33
win rate35.29
sample136
mean trade pct-0.29
average win pct1
average loss pct0.99
CI low pct-0.6
CI high pct0.03
quarter Kelly0
cost bps side6
calendar windows34