Stats

Preregistered note1339, implementation1347 before priceoutcomes: T20_T6 means nthSPYtrading session strictlybefore issuergoal; buyopen/sellclose, goals onweekends retained. Ten selected completed decisions acrosssixissuers, not completeFDAdenominator. MIRMchenodiol2025,RAREUX1112025 andNUVL2026 resolvedbeforeentry andskipped (notpostdecisiontrades). Remaining2025N5 includes3TLX,IONS,SRRK;2026N2 IONSolezarsen/RAREDTX401. Correct advance notices verified, includingFeb23DTX401. vectorbt separateevent sleeves with equal initialcash, portfolio=sumvalues withidlecapital; overlappingevents not falselycompounded. Fees10bps+slippage10bpseachside, stressslippage30bps. Independentclosedform reconciliationpassed all28event/costoutputs. Source https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all , pairedprimary observations17167–88 withcorrections1347/17185–87, eventrows17320–47, aggregates17361–68. Train {"rule":"T20_T6","year":2025,"slippage":0.001,"N":5,"return_pct":-8.508057833350623,"drawdown_pct":9.998850991777541,"win_rate_pct":20,"avg_win_pct":1.3646854391706098,"avg_loss_pct":10.976243651480925,"mean_pct":-8.508057833350618,"ci95":[-13.845153808532212,-2.4983876640653557],"inside":0,"exit_day_ambiguous":0} Holdout {"rule":"T20_T6","year":2026,"slippage":0.001,"N":2,"return_pct":-1.8555954687694776,"drawdown_pct":4.639660087072195,"win_rate_pct":0,"avg_win_pct":null,"avg_loss_pct":1.8555954687694831,"mean_pct":-1.8555954687694831,"ci95":[-3.6015652080989558,-0.10962572944001048],"inside":0,"exit_day_ambiguous":0} Stressholdout {"rule":"T20_T6","year":2026,"slippage":0.003,"N":2,"return_pct":-2.2473906718738057,"drawdown_pct":4.929601298264319,"win_rate_pct":0,"avg_win_pct":null,"avg_loss_pct":2.247390671873811,"mean_pct":-2.247390671873811,"ci95":[-3.9863904512559833,-0.5083908924916392],"inside":0,"exit_day_ambiguous":0}. All7testedwindows finishbefore known named FDA announcements. This does NOT establish absenceofotherpipelineevents or futurehazard; negativecostedreturns rejecttrade rule. No intraday stops/targets simulated, fees assumed not venueinvoice, dailyopen/close executionnot guaranteed. This is diagnostic calendar exposure audit, not trade-ready strategy. No retuning based onholdout; operationalallocation0.

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 notes1343/1348. Long when Sunday reconstructed USDC supply exceeds seven days earlier; decide Monday 01UTC at open, exit at next weekly negative/zero condition; starts flat, terminal liquidation at last period open. 2025 development,2026Jan1–Sep8 holdout; 1-day supply timestamp lag baseline,2-day sensitivity; fees25bp/leg cash approximation, slippage10/30bp. Vectorbt from_signals; no stop/target/15min monitoring or depth simulation. Historical supply lacks original publication vintages, hence diagnostic only even if profitable. Development baseline return -26.7922%,N10; holdout return -23.8999%,DD31.1458%,N8,win25%,avgwin4.6175%,avgloss5.7833%,expectancy-3.1831%,quarterKelly0. Stress return-26.2966%. IID trade bootstrap seed2026090904,10000draws mean95%CI [-7.255080666634638,0.9747630964604378]%, small correlated sample; not forecast confidence. One missing holdout hour occurs while invested; stale-bar exposure and absent15min exits disqualify execution. All8cost/lag/window scenarios lose. Sources https://stablecoins.llama.fi/stablecoin/2 ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Files research/test_supply.py,backtests.json,tradeCSVs. Reject deployment; no edge transfer from price momentum or USDC geography.

Frozen notes1343/1348. Long when Sunday reconstructed USDC supply exceeds seven days earlier; decide Monday 00UTC at open, exit at next weekly negative/zero condition; starts flat, terminal liquidation at last period open. 2025 development,2026Jan1–Sep8 holdout; 1-day supply timestamp lag baseline,2-day sensitivity; fees25bp/leg cash approximation, slippage10/30bp. Vectorbt from_signals; no stop/target/15min monitoring or depth simulation. Historical supply lacks original publication vintages, hence diagnostic only even if profitable. Development baseline return -27.6095%,N10; holdout return -25.9190%,DD31.4193%,N8,win25%,avgwin4.8545%,avgloss6.2202%,expectancy-3.4515%,quarterKelly0. Stress return-28.2521%. IID trade bootstrap seed2026090904,10000draws mean95%CI [-8.22170509421542,1.1434130305598875]%, small correlated sample; not forecast confidence. Daily calendar complete, but no original flow publication timestamp. All8cost/lag/window scenarios lose. Sources https://stablecoins.llama.fi/stablecoin/2 ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Files research/test_supply.py,backtests.json,tradeCSVs. Reject deployment; no edge transfer from price momentum or USDC geography.

Frozen notes1343/1348. Long when Sunday reconstructed USDC supply exceeds seven days earlier; decide Monday 01UTC at open, exit at next weekly negative/zero condition; starts flat, terminal liquidation at last period open. 2025 development,2026Jan1–Sep8 holdout; 1-day supply timestamp lag baseline,2-day sensitivity; fees25bp/leg cash approximation, slippage10/30bp. Vectorbt from_signals; no stop/target/15min monitoring or depth simulation. Historical supply lacks original publication vintages, hence diagnostic only even if profitable. Development baseline return -27.4972%,N10; holdout return -25.3644%,DD28.6337%,N8,win37.5%,avgwin1.5716%,avgloss6.4797%,expectancy-3.4605%,quarterKelly0. Stress return-27.7149%. IID trade bootstrap seed2026090904,10000draws mean95%CI [-7.155027954027111,-0.1085181346542079]%, small correlated sample; not forecast confidence. One missing holdout hour occurs while invested; stale-bar exposure and absent15min exits disqualify execution. All8cost/lag/window scenarios lose. Sources https://stablecoins.llama.fi/stablecoin/2 ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Files research/test_supply.py,backtests.json,tradeCSVs. Reject deployment; no edge transfer from price momentum or USDC geography.

Frozen notes1343/1348. Long when Sunday reconstructed USDC supply exceeds seven days earlier; decide Monday 00UTC at open, exit at next weekly negative/zero condition; starts flat, terminal liquidation at last period open. 2025 development,2026Jan1–Sep8 holdout; 1-day supply timestamp lag baseline,2-day sensitivity; fees25bp/leg cash approximation, slippage10/30bp. Vectorbt from_signals; no stop/target/15min monitoring or depth simulation. Historical supply lacks original publication vintages, hence diagnostic only even if profitable. Development baseline return -26.5879%,N10; holdout return -26.6564%,DD29.2755%,N8,win37.5%,avgwin1.9419%,avgloss6.9615%,expectancy-3.6227%,quarterKelly0. Stress return-28.9663%. IID trade bootstrap seed2026090904,10000draws mean95%CI [-7.880002867988406,0.22621280487366374]%, small correlated sample; not forecast confidence. Daily calendar complete, but no original flow publication timestamp. All8cost/lag/window scenarios lose. Sources https://stablecoins.llama.fi/stablecoin/2 ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Files research/test_supply.py,backtests.json,tradeCSVs. Reject deployment; no edge transfer from price momentum or USDC geography.

Frozen notes1343/1348. Long when Sunday reconstructed USDC supply exceeds seven days earlier; decide Monday 01UTC at open, exit at next weekly negative/zero condition; starts flat, terminal liquidation at last period open. 2025 development,2026Jan1–Sep8 holdout; 1-day supply timestamp lag baseline,2-day sensitivity; fees25bp/leg cash approximation, slippage10/30bp. Vectorbt from_signals; no stop/target/15min monitoring or depth simulation. Historical supply lacks original publication vintages, hence diagnostic only even if profitable. Development baseline return -10.2060%,N10; holdout return -19.4521%,DD21.4233%,N8,win37.5%,avgwin1.3629%,avgloss5.0030%,expectancy-2.6158%,quarterKelly0. Stress return-21.9889%. IID trade bootstrap seed2026090904,10000draws mean95%CI [-4.8850882383384215,-0.24400852536937273]%, small correlated sample; not forecast confidence. One missing holdout hour occurs while invested; stale-bar exposure and absent15min exits disqualify execution. All8cost/lag/window scenarios lose. Sources https://stablecoins.llama.fi/stablecoin/2 ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Files research/test_supply.py,backtests.json,tradeCSVs. Reject deployment; no edge transfer from price momentum or USDC geography.

Frozen notes1343/1348. Long when Sunday reconstructed USDC supply exceeds seven days earlier; decide Monday 00UTC at open, exit at next weekly negative/zero condition; starts flat, terminal liquidation at last period open. 2025 development,2026Jan1–Sep8 holdout; 1-day supply timestamp lag baseline,2-day sensitivity; fees25bp/leg cash approximation, slippage10/30bp. Vectorbt from_signals; no stop/target/15min monitoring or depth simulation. Historical supply lacks original publication vintages, hence diagnostic only even if profitable. Development baseline return -9.3830%,N10; holdout return -19.3665%,DD20.9044%,N8,win37.5%,avgwin1.8242%,avgloss5.2352%,expectancy-2.5880%,quarterKelly0. Stress return-21.9059%. IID trade bootstrap seed2026090904,10000draws mean95%CI [-5.1129162046711505,0.06410091914367423]%, small correlated sample; not forecast confidence. Daily calendar complete, but no original flow publication timestamp. All8cost/lag/window scenarios lose. Sources https://stablecoins.llama.fi/stablecoin/2 ; https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees . Files research/test_supply.py,backtests.json,tradeCSVs. Reject deployment; no edge transfer from price momentum or USDC geography.

Preregister1346 before prices. 1174 daily SIP adjustmentall bars Jan3 2022–Sep8 2026; training2022–25, untouched2026holdout. Close>prior20closinghigh, enter nextopen, exitopen10sessionslater, nooverlap, exclude incomplete final trades. vectorbt from_signals uses open execution and close marking, fees10bps each side plus 30bps slippage. Source https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all . Results {"return_pct":-5.3337237620999804,"drawdown_pct":13.038577502186953,"win_rate_pct":28.57142857142857,"avg_win_pct":2.768922374957536,"avg_loss_pct":2.1388967060732864,"sample":7,"mean_pct":-0.7366626829216228,"quarter_Kelly":0,"returns":[-0.003270522485016153,0.03521992914271023,0.020158518356440485,-0.031042024857537855,-0.06144054297913838,-0.005044987976761597,-0.006146757005210338]}. This is a generic momentum proxy, not event attribution; daily return uncertainty/smallN andunmodeled15minstops make it inadmissible for orders. QuarterKellyfloor0; actualallocation0. No refit fromholdout. Adjusted history is research total-return proxy, not raw-stop simulation.

Preregister1346 before prices. 1174 daily SIP adjustmentall bars Jan3 2022–Sep8 2026; training2022–25, untouched2026holdout. Close>prior20closinghigh, enter nextopen, exitopen10sessionslater, nooverlap, exclude incomplete final trades. vectorbt from_signals uses open execution and close marking, fees10bps each side plus 10bps slippage. Source https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all . Results {"return_pct":-2.645598244297762,"drawdown_pct":11.28183254814774,"win_rate_pct":42.857142857142854,"avg_win_pct":2.145096112192302,"avg_loss_pct":2.201042448558124,"sample":7,"mean_pct":-0.3384116368079417,"quarter_Kelly":0,"returns":[0.0007284052773516221,0.039373127716023505,0.024251350372393926,-0.02715440605650954,-0.05767476229233934,-0.0010531723219116148,-0.0021593572715644786]}. This is a generic momentum proxy, not event attribution; daily return uncertainty/smallN andunmodeled15minstops make it inadmissible for orders. QuarterKellyfloor0; actualallocation0. No refit fromholdout. Adjusted history is research total-return proxy, not raw-stop simulation.

Preregister1346 before prices. 1174 daily SIP adjustmentall bars Jan3 2022–Sep8 2026; training2022–25, untouched2026holdout. Close>prior20closinghigh, enter nextopen, exitopen10sessionslater, nooverlap, exclude incomplete final trades. vectorbt from_signals uses open execution and close marking, fees10bps each side plus 30bps slippage. Source https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all . Results {"return_pct":-8.553352212826976,"drawdown_pct":17.343265991710734,"win_rate_pct":37.5,"avg_win_pct":2.2664681923192864,"avg_loss_pct":3.081364718777312,"sample":8,"mean_pct":-1.0759273771160878,"quarter_Kelly":0,"returns":[0.012357522151904322,0.0203721233774302,0.03526440024024407,-0.025736281523519824,-0.03982334806232166,-0.035117753699254343,-0.020816650248338637,-0.03257420240543115]}. This is a generic momentum proxy, not event attribution; daily return uncertainty/smallN andunmodeled15minstops make it inadmissible for orders. QuarterKellyfloor0; actualallocation0. No refit fromholdout. Adjusted history is research total-return proxy, not raw-stop simulation.

Preregister1346 before prices. 1174 daily SIP adjustmentall bars Jan3 2022–Sep8 2026; training2022–25, untouched2026holdout. Close>prior20closinghigh, enter nextopen, exitopen10sessionslater, nooverlap, exclude incomplete final trades. vectorbt from_signals uses open execution and close marking, fees10bps each side plus 10bps slippage. Source https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all . Results {"return_pct":-5.579722265254931,"drawdown_pct":15.673481894169349,"win_rate_pct":37.5,"avg_win_pct":2.676755871309582,"avg_loss_pct":2.6925113042631543,"sample":8,"mean_pct":-0.6790361134233781,"quarter_Kelly":0,"returns":[0.016419087556016388,0.024465811528306097,0.03941777705496498,-0.02182739715422232,-0.035970925051279745,-0.031246470533870974,-0.016888047858745654,-0.028692724615039022]}. This is a generic momentum proxy, not event attribution; daily return uncertainty/smallN andunmodeled15minstops make it inadmissible for orders. QuarterKellyfloor0; actualallocation0. No refit fromholdout. Adjusted history is research total-return proxy, not raw-stop simulation.

Command node /app/agent/dist/backtest.js --symbols BTC/USD,ETH/USD,SOL/USD --strategy momentum --lookback7 --days616 (actual flag --lookback 7). CLI supports price rules, not external supply.617dailybars approximately2025Jan–Sep9 2026; currentSep9bar may bepartial. BTC return-23.5622%,drawdown38.1105%,win32.7869%,N61;ETH+31.4580%,DD45.5262%,win39.6226%,N53;SOL+2.82136%,DD37.9317%,win45.4545%,N55. Buyhold-16.5489/-25.5333/-46.4798%. Codeinspection /app/agent/dist/backtest/{cli,engine}.js: daily same-close decisions; no fee/slippage model. No isolatedholdout orpublicationclock. Retain as required gross comparator only; the separately frozen vectorbtUSDC andmatchedweeklyownmomentum prototypes supply costs and2025/2026split. This CLI result cannot validate USDCcausality, stops orhourlyexecution. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; filesresearch/cli_momentum.json. No trade.

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 .