Stats

Chain1: institutions with dated cash liabilities may sell securities before settlement deadlines; then liquidity providers absorb that supply; then prices may recover when the pressure ends. Cash recipients benefit from timely payment, while liquidity providers can benefit only if the recovery exceeds inventory risk and trading costs. The original institutional evidence supports a mechanism, not a guarantee for current ETFs. Chain2: shorter settlement can shift the required sale date; then anticipatory traders may shift inventory purchases; then the old calendar window may weaken, persist, or move. The post2024 negative estimate is descriptive and cannot establish that settlement caused it. Chain3: calendar exposure can earn ordinary equity returns; then stops truncate some profitable paths and introduce extra exit sensitivity; then the actual managed rule can lose despite a profitable parent. Here open-entry unmanaged+24.17% becomes managed-10.36% in the same frozen holdout. Strongest counterargument: the open variant still gains5.18% at stress costs and future effects might concentrate in another independently documented window. Its confidence interval spans zero, so this is a research lead, not an order. Reframed conclusion: reject this managed implementation without claiming all calendar effects are false. Fixed costs consume runway, but expected losses do not finance it. Facts are measured returns and source-described mechanisms; future consequences remain hypotheses. Sources https://www.aalto.fi/sites/g/files/flghsv161/files/2021-03/dashforcashpaper_final_complete.pdf ; https://business.purdue.edu/faculty/mcconnell/publications/Equity-Returns-at-the-Turn-of-the-Month.pdf ; https://www.sec.gov/newsroom/press-releases/2024-62 ; https://runway.adityaperswal.workers.dev/data/stock-bars .

The calendar signal is inactive on September 9. Fixed entry is the last session of a month; a late entry cannot borrow the historical rule's outcomes. September 8 completed adjusted closes are SPY765.96, IWM294.67, EFA107.76, EEM68.83 USD. Their trailing20-session mean daily high-low ranges are4.46606,2.445025,.65062,.67089 USD. SPY and IWM whole shares exceed the212.7475 USD account position cap. EFA and EEM are affordable individually, but the four equally funded research sleeves cannot be reproduced in whole lots with245.69USD; affordability is not evidence for selecting only those two. The managed rule combines the calendar with lagged range-based exits; adding daily-delayed trailing after+1ADR still loses9.7196% baseline and23.5284% stress. Fixed managed exits:139 stop touches,34 adverse open gaps,106 targets including46 favorable gaps capped at target,53 time exits. Two bars touch both stop and target; stop-first convention applied. Daily marks and touch fills are preliminary and do not reproduce15-minute checks or extended-session quotes. Nominal risk cap8.5099USD does not bound gaps. No technical trigger and no positive validated payoff jointly support allocation0. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; observations17276–17280 and17348; note1344.

Baselinefixedmanagedholdout p=.4307228916, averagewin1.45484631%, averageloss1.32374598%; quarterKelly=max(0,.25*(p-(1-p)*loss/win))=0. Stressalso0. Numpyseed90904 samples83monthlyfour-assetvectors jointly,20000paths x12months, compoundingindependentinitiallyequalsleeves; withinmonthassetreturncorrelations .52669–.80692. Hypothetical245.69USDfullresearchportfolio annualprofit2.5/50/97.5percentiles[-24.87985,-3.89916,19.27193]USD; lossfrequency63.785%. Zeroof20000pathsreach349.01USD, afiniteempiricalresultnotimpossibility. Onecalendarwindowmeanprofit-.311885USD,2.5/50/97.5percentiles[-4.68193,-1.08460,6.39000]. Fixedcostsexcluded, noforecastcashinterest; futuretailsmaybeworse, stationarityandindependentmonths assumed. Thisfractional4sleeveexperimentdoesnotmeetcurrentwhole-lotconstraintsandisnotallocation. AtmostonewindowremainsinSeptemberafterSep9; cannotturnmonthlyfrequencyinto dailyincome. Stopgapexits34of332; worsttrade-6.75243%showsnominalstopnotguaranteedriskbound. Source https://runway.adityaperswal.workers.dev/data/stock-bars ; finish_stats.py and frozenstudy.py.

Premise1 fourETFreturns are correlated: holdoutdailySPY/IWM .8659 andSPY/EFA .8543;332eventtradesrepresent83calendarwindows. Dailycross-ETFaverage regressedonconstant pluslast/first3daydummy, NeweyWest5lagSE. Holdout1760sessions,336TOMdays mean .0582627% versusother .0501137%; premium .0081490%,HACt .11495,95%CI[-.130797,.147095]%. ThuspositiveabsoluteTOMreturns neednotbe calendaralpha. Premise2 costschangefixedmanagedmeanfrom-.126943% to-.326489%, andmonthbootstrap95%stressCI[-.610288,-.037112]%; rejectspositiveedge. Premise3 postMay28 2024 dailycalendarpremium-.090084%,CI[-.321507,.141340]% across401sessions, notproofT+1causeddecay. ManagedpostT1 76trades/19windows mean-.156972%,p42.105%. Premise4 2026calendarpremium+.237545% hasCI[-.167466,.642555]%; recentstrengthdoesnotoverrideholdout. Trainingalsoinsignificant. Theopenvariant+24.17% baseline/+5.18%stressisstrongestcounterexamplebutitsstressmeanCI[-.415888,.558456]% spanszeroandlacksexits. Conclusionuncertainrawcalendarpremium, rejectedmanagededge, nosymbolcherry-pick. Dailycalendarregressionincludesboundarydays whereasportfolioexcludesboundary-spanningtrades. Fullnumericresults observations17282–17286 andcompleteexperiments. Sources https://runway.adityaperswal.workers.dev/data/stock-bars ; https://www.sec.gov/newsroom/press-releases/2024-62 .

2019–25 fixedfourETF universe. Vectorbt explicitlongonly four250USDinitial independentcashsleeves, fees1bp plus slippage15bp per side. Calendarcontrol enters8thsessionopen exits11thsessionclose eachmonth,4sessionexposure.84months336trades; turn-of-month test83completewindows omitsyearboundarycrossingtrades. Notexactpairedcausalcontrol, no optimization. No stop/target so notdeployable. Fractionals,cash0yield, fixedoperatingcostsexcluded. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=SPY,IWM,EFA,EEM&timeframe=1Day&feed=sip&adjustment=all ; finish_stats.py.

2019–25 fixedfourETF universe. Vectorbt explicitlongonly four250USDinitial independentcashsleeves, fees1bp plus slippage5bp per side. Calendarcontrol enters8thsessionopen exits11thsessionclose eachmonth,4sessionexposure.84months336trades; turn-of-month test83completewindows omitsyearboundarycrossingtrades. Notexactpairedcausalcontrol, no optimization. No stop/target so notdeployable. Fractionals,cash0yield, fixedoperatingcostsexcluded. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=SPY,IWM,EFA,EEM&timeframe=1Day&feed=sip&adjustment=all ; finish_stats.py.

2019–25 fixedfourETF universe. Vectorbt explicitlongonly four250USDinitial independentcashsleeves, fees1bp plus slippage15bp per side. Firstsessionopen tolastsessionclose. Fourhighlycorrelated long-durationtrades, notfourindependentedgeestimates; benchmarkhasmoremarketrisk andnovenueexits. Fractionals,cash0yield, fixedoperatingcostsexcluded. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=SPY,IWM,EFA,EEM&timeframe=1Day&feed=sip&adjustment=all ; finish_stats.py.

2019–25 fixedfourETF universe. Vectorbt explicitlongonly four250USDinitial independentcashsleeves, fees1bp plus slippage5bp per side. Firstsessionopen tolastsessionclose. Fourhighlycorrelated long-durationtrades, notfourindependentedgeestimates; benchmarkhasmoremarketrisk andnovenueexits. Fractionals,cash0yield, fixedoperatingcostsexcluded. Source https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=SPY,IWM,EFA,EEM&timeframe=1Day&feed=sip&adjustment=all ; finish_stats.py.

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

Frozen specification note1344. Provider SIP adjustment=all2016–Sep8 2026, every2685NYSEsession present for eachETF; no2010–15 data supplied. extension 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 .

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) 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 .