A frozen note, number 2174, outlines the premises of a test. The test uses nine sector ETFs, with daily bars adjusted for dividends and splits, from January 4, 2021, at 05:00:00+00:00 to September 10, 2026, at 04:00:00+00:00. The test buys the worst-performing sector from the previous five sessions if it is negative, using the previous close as a signal to fill at the next open, with one cash-only fractional sleeve. The test uses a stop loss based on the average true range, or ATR, which is a measure of volatility, set at 20 periods and 1 times the range, with a target of 3 times the range and a 2 times the range quarter, and a breakeven point the following day. The test also uses gap open and conservative barriers, with a weekly time exit. The test only enters a trade if the prior 20-day realized volatility of the SPY, which is a measure of how much the price of the SPY has fluctuated, is greater than 20%, but not based on the VIX, which is a measure of expected volatility. The fees for the test are 1 basis point, or 0.01%, plus 15 basis points, or 0.15%, for slippage on each side. The test has four phases per day, using a vectorized backtesting approach, with long-only positions and cash sharing, reconciled every day, with no hidden leverage. The mean trade confidence interval, which is a measure of the range of possible values, is between -1.202347268207001% and 0.9499598378533135% at a 95% confidence level, based on 5000 bootstrap samples clustered by month. The reasons for trade exits are: 19 times due to time, 17 times due to stop loss, 3 times due to gap stop, and 2 times due to target, with 3 partial exits and 3 same-day final exits. The current incomplete week is excluded, and all observations and source data are archived. An audit of opening partials before intraday stops did not change any cells. The conclusion of the test is that the primary strategy fails to be robust to costs, and controls cannot replace the primary strategy after holdout inspection. The mechanics of adjusted dividends, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not based on Nagel's 48-industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a trading strategy. The strategy involves nine sector ETFs, with daily bars adjusted for splits and dividends, from January 4, 2016, to December 31, 2020. Each week, the strategy buys the sector with the worst return over the prior five sessions, but only if that return is negative. The buy is executed at the next open, with a signal from the prior close, and is a cash-only, fractional sleeve. The strategy uses a stop loss based on the average true range over 20 days, with a target of three times the risk and a barrier of two times the risk, and exits at the end of the week or when the target is reached. The strategy also considers the realized volatility of the SPY over the prior 20 days, but not the VIX. Fees are one basis point plus 15 basis points for slippage on each side. The strategy has four phases per day and is reconciled daily. The mean trade confidence interval is between -0.7943843415420596% and 0.8182585213220581% over 5000 bootstrap samples. The reasons for trades are: stop losses 19 times, time exits 23 times, and target reaches 1 time. There are 2 partial exits and 7 same-day final exits. The current incomplete week is excluded from the analysis, and all observations and source data are archived. An audit of the opening partial before intraday stop did not change any results. The conclusion is that the primary strategy fails to be robust, and controls cannot replace the primary strategy after inspection. Adjusted dividend mechanics, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not related to Nagel's 48 industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a test. The test uses nine sector ETFs with daily bars adjusted for dividends and splits from January 4, 2021, at 05:00:00+00:00 to September 10, 2026, at 04:00:00+00:00. The test buys the worst-performing sector from the previous five sessions if it was negative, using the previous close as a signal to fill the next open, with one cash-only fractional sleeve. The test uses a stop loss of one times the average true range over 20 days, a target of three times the average true range, and a breakeven point the following day. It also uses gap open, conservative barriers, and a weekly time exit. The test only occurs if the realized volatility of the SPY over the previous 20 days is greater than 20%, and it does not use the VIX. The fees are one basis point plus slippage of five basis points on each side, with four phases per day and vector backtesting for long-only cash sharing, reconciled daily, with no hidden leverage. The mean trade confidence interval at 95% is between -1.034219621696518% and 1.1220167616599663%, with a month cluster of 5000 bootstrap samples. The reasons for trade exits are: time, 19; stop, 17; gap stop, 3; and target, 2, with three partial exits and three same-day final exits. The current incomplete week is excluded, and all observations and source data are archived. An audit of opening partials before intraday stops did not change any cells. The conclusion of this research-only test is that the primary strategy fails to be robust to costs, and controls cannot replace the primary strategy after holdout inspection. The mechanics of adjusted dividends, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not Nagel's 48-industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a trading strategy. The strategy involves nine sector ETFs, with daily bars adjusted for dividends and splits, from January 4, 2016, to December 31, 2020. Each week, the strategy buys the sector with the worst return over the prior five sessions, but only if that return is negative. The buy is executed at the next open, with a signal from the prior close. The strategy uses a cash-only approach, with fractional shares allowed. A stop-loss is set at one times the average true range over 20 days, with a target of three times that range and a maximum gain of two times that range per quarter. The strategy also includes a barrier to prevent large losses and a weekly time exit. A volatility gate is used, based on the prior 20-day realized volatility of the SPY ETF, but not the VIX index. The strategy has fees of one basis point plus slippage of five basis points on each side. There are four phases per day, with vectorized backtesting and long-only cash sharing, reconciled daily. No hidden leverage is used. The mean trade confidence interval is between -0.64% and 0.99%, based on 5000 bootstrap samples. The reasons for trades are: gap stop 2 times, time exit 23 times, stop-loss 17 times, and target reached 1 time. There were 3 partial exits and 5 same-day final exits. The current incomplete week is excluded from the analysis, and all observations and source data are archived. An audit of opening partials before intraday stops did not change any results. The conclusion is that the primary strategy fails to be robust, and controls cannot replace the primary strategy after inspection. Certain mechanics, such as adjusted dividend handling and slippage estimates, remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This strategy is not related to Nagel's 48-industry daily short portfolio.
A news scan found the note: Frozen note2174. The test was based on 9 sector ETFs, with daily price changes adjusted for dividends and splits, from January 4, 2021, 05:00:00+00:00 to September 10, 2026, 04:00:00+00:00. The test bought the worst performing sector of the previous 10 sessions if it was negative, and sold at the next open. The test used a cash-only strategy, with a stop loss of 1 times the 20-day average true range, a target of 3 times the true range, and a 2 true range quarter. The test also used a weekly time exit and conservative barriers. The fees were 1 basis point plus 15 basis points for slippage on each side. The test was reconciled every day, with no hidden leverage. The mean trade confidence interval was between -0.31033285590242954% and 0.26975274239519337%. The reasons for the trades were: gap target 1, stop 107, time 133, target 4, and gap stop 13. There were 35 partial trades and 23 same-day final exits. The current incomplete week was excluded, and all observations and source data were archived. An audit of the opening partial before intraday stop did not change any results. The conclusion is that the primary test failed due to cost robustness, and controls cannot replace the primary test after holdout inspection. The mechanics of adjusted dividends, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This test is not related to Nagel's 48 industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a trading strategy. The strategy involves nine sector ETFs, with daily bars adjusted for dividends and splits, from January 4, 2016, to December 31, 2020. Each week, the strategy buys the sector with the worst return over the prior 10 sessions, but only if that return is negative. The buy order is filled at the next open, with a signal from the prior close, and is a cash-only, fractional sleeve, meaning it can buy a fraction of a share. The strategy uses a stop-loss, which is a price level at which to sell a security, set at 1 times the average true range, or ATR, over 20 sessions, and a target price set at 3 times the ATR, with an additional 2 times the ATR as a quarterly target, and a breakeven point the following day. The strategy also includes gap open, conservative barriers, and a weekly time exit, meaning it will sell at the end of the week. There are fees of 1 basis point, plus slippage of 15 basis points on each side of the trade. The strategy has four phases per day, uses vectorized backtesting, and is long-only, meaning it only buys securities, with cash sharing reconciled daily. The strategy does not use hidden leverage, meaning it does not use borrowed money to buy more securities. The mean trade confidence interval, a measure of the strategy's performance, is between -0.3248899491009365 and 0.09155388045638681 percent, based on 5000 bootstrap samples, which is a statistical method. The reasons for the trades are: 113 times due to time, 7 times due to a gap stop, 98 times due to a stop, 2 times due to a gap target, and 1 time due to a target. There were 18 partial trades and 33 same-day final exits. The current, incomplete week is excluded from the analysis, and all observations and source data are archived. An audit of the opening partial, before intraday stop, did not change any cells. The conclusion, for research purposes only, is that the primary strategy fails to be robust, and controls cannot replace the primary strategy after holdout inspection. The mechanics of adjusted dividends, actual 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not Nagel's 48-industry daily short portfolio.
A frozen note, number 2174, was reviewed. The test was based on 9 sector ETFs, with daily price changes adjusted for dividends and splits, from January 4, 2021, to September 10, 2026. The test bought the worst performing sector from the previous 10 sessions, but only if it had a negative return, and sold at the next open. The test used a stop loss, which is a price level at which to automatically sell a security, set at 1 times the average true range, or ATR, over 20 days, and a target price set at 3 times the ATR. The test also used a barrier, which is a price level that triggers a sale, set at 2 times the ATR above the target, and exited positions at the end of each week. There were fees of 1 basis point, or 0.01%, plus slippage, which is the difference between the expected and actual price of a trade, of 5 basis points, on both the buy and sell sides. The test was run in four phases per day, using a vectorized backtesting framework, and was long only, meaning it only bought securities, and used cash sharing, which is a method of allocating cash to different securities. The test was reconciled daily, and there was no hidden leverage, which is borrowing to invest. The average trade return was between -0.12% and 0.50% with 95% confidence. The test had 41 partial exits, which are sales of part of a position, and 23 same-day final exits. The current incomplete week was excluded from the test, and all observations and source data were archived. An audit of the opening partial before intraday stop did not change any results. The conclusion of the test was that the primary strategy failed to be robust, and the controls, which are tests used for comparison, could not replace the primary strategy after inspection. The test did not validate dividend mechanics, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader, or PDT, rules. The source of the data was https://runway.adityaperswal.workers.dev/data/stock-bars. This test is not related to Nagel's 48 industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a trading strategy. The strategy involves nine sector ETFs, with daily bars adjusted for dividends and splits, from January 4, 2016, to December 31, 2020. Each week, the strategy buys the sector with the worst return over the previous 10 sessions, but only if that return is negative. The buy is executed at the next open, with a signal from the prior close. The strategy uses a cash-only approach, with fractional shares allowed. A stop-loss is set at 1 times the average true range over 20 days, with a target of 3 times the average true range, and an additional 2 times the average true range for a quarter, and a breakeven point the following day. The strategy also includes gap open, conservative barriers, and a weekly time exit. Fees are 1 basis point, plus 5 basis points for slippage on each side. The strategy has four phases per day, with vectorized backtesting, long-only positions, and cash sharing, reconciled daily. There is no hidden leverage. The mean trade confidence interval is between -0.0856% and 0.3319%, based on 5000 bootstrap samples. The reasons for trades include time, gap stop, stop, gap target, and target, with 23 partial trades and 24 same-day final exits. The current incomplete week is excluded, and all observations and source data are archived. An audit of opening partials before intraday stops did not change any results. The conclusion is that the primary strategy fails to provide cost robustness, and controls cannot replace the primary strategy after inspection. Adjusted dividend mechanics, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not related to Nagel's 48-industry daily short portfolio.
A news scan found the note about the REV9110400 neighbor3_15_test, which costed a weekly reversal. The note starts with the premises that it uses nine sector ETFs and daily bars that are all-adjusted for SIP, from January 4, 2021, at 05:00:00+00:00 to September 10, 2026, at 04:00:00+00:00. The first session of each week buys the sector with the worst return over the prior three sessions, but only if that return is negative. It uses a signal from the prior close to fill at the next open, and it only uses cash, with fractional shares allowed in one sleeve. The note mentions an ATR20 stop, which is a stop-loss order based on the average true range over 20 periods, and a target of 3R, which means three times the risk, with a 2R quarter, which is two times the risk for a quarter of the trade, and a break-even point the following day. It also mentions a gap open, conservative barriers, and a weekly time exit, which means the trade is closed at the end of the week. The fees are 1 basis point plus 15 basis points for slippage on each side, which means the difference between the expected price and the actual price. The note also mentions four phases per day, vectorbt long-only cash sharing reconciled every day, and no hidden leverage, which means borrowing money to invest. The mean trade confidence interval at 95% is between -0.41598039756013533% and 0.1716584000385664%, which is a measure of the uncertainty of the trade returns, based on 5000 bootstrap samples, which is a statistical method to estimate the variability of the returns. The reasons for the trades are given as a dictionary with keys for stop, gap stop, time, target, and gap target, with values 118, 15, 120, 7, and 1, respectively. The note also mentions partials, which are 34, and same-day final exits, which are also 34. The current incomplete week is excluded, and all observations and source data are archived. An audit of the opening partial before the intraday stop did not change any cell. The conclusion is that the primary strategy fails the cost robustness test, and the controls cannot replace the primary strategy after holdout inspection. The note also mentions that adjusted dividend mechanics, actual 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and PDT, which is a pattern day trader rule, remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not Nagel's 48 industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a trading strategy. The strategy involves nine sector ETFs, with daily bars adjusted for dividends and splits, from January 4, 2016, to December 31, 2020. The strategy buys the worst-performing sector from the previous three sessions if it is negative, using the previous close as a signal to fill the next open, with one cash-only fractional sleeve. The strategy uses a stop loss of 1 times the 20-day average true range, a target of 3 times the range, and a 2 times range quarter, with a barrier to prevent large losses, and exits at the end of the week. The strategy also includes fees of 1 basis point plus 15 basis points for slippage on each side, with four phases per day, using a vectorized backtesting approach, with long-only positions, cash sharing, and reconciliation every day, with no hidden leverage. The mean trade confidence interval is between -0.28% and 0.31%, with a monthly cluster of 5000 bootstrap samples. The reasons for trade exits are: 94 stops, 108 time exits, 5 target exits, 13 gap stops, and 4 gap targets, with 26 partial exits and 32 same-day final exits. The current incomplete week is excluded, and all observations and sources are archived, with an audit of opening partials before intraday stops that did not change any cells. The conclusion is that the primary strategy fails to be robust, and controls cannot replace the primary strategy after inspection. Adjusted dividend mechanics, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and this adaptation is not related to Nagel's 48 industry daily short portfolio.
A news scan found the note: Frozen note2174. The premises of this test are based on 9 sector ETFs, with all-adjusted daily bars from 2021-01-04 05:00:00+00:00 through 2026-09-10 04:00:00+00:00. The strategy buys the worst-performing sector from the previous 3 sessions if it is negative, using the prior close as a signal to fill the next open, with one cash-only fractional sleeve. The strategy uses a 20-day average true range, or ATR, which is a measure of volatility, as a stop loss, with a target of 3 times the risk, or R, and a 2R quarter, and a break-even point, or BE, the following day. It also uses a gap open, conservative barriers, and a weekly time exit. The fees are 1 basis point, or bp, plus 5bp for slippage, which is the difference between the expected and actual price of a trade, on each side. The strategy is reconciled every day, with four phases per day, using a vectorbt, which is a backtesting framework, long-only, cash-sharing approach, with no hidden leverage, which is the use of borrowed money to increase potential returns. The mean trade confidence interval, or CI, is between -0.2562198984201958% and 0.33990779442520025%, with a month cluster of 5000 bootstrap, which is a statistical method. The reasons for the trades are: 114 stops, 14 gap stops, 125 time exits, 7 targets, and 1 gap target. There were 36 partial trades and 32 same-day final exits. The current incomplete week is excluded, and all observations and source data are archived. An audit of the opening partial before intraday stop did not change any cell. The conclusion of this research is that the primary strategy fails to be robust, and the controls cannot replace the primary strategy after holdout inspection, which is a method of testing a strategy. The mechanics of adjusted dividends, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader, or PDT, rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not based on Nagel's 48 industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a trading strategy. The strategy involves nine sector ETFs, with daily price adjustments, from January 4, 2016, to December 31, 2020. Each week, the strategy buys the sector that performed the worst over the previous three sessions, but only if it had a negative return. The purchase is made at the next open price, with a signal from the previous close. The strategy uses a cash-only approach, with fractional shares allowed. A stop-loss is set at one times the average true range over 20 days, with a target of three times the range and an additional two times the range as a quarterly benchmark. The strategy also includes a gap open and conservative barriers, with a weekly time exit. Fees are one basis point plus slippage of five basis points on each side. The strategy has four phases per day, with vector-based backtesting and long-only cash sharing, reconciled daily. There is no hidden leverage. The mean trade confidence interval is between -0.0623% and 0.5203%, based on 5000 bootstrap samples. The reasons for trade exits are: gap stop 14 times, time exit 117 times, stop loss 84 times, target reached 5 times, and gap target 4 times. There were 30 partial exits and 28 same-day final exits. The current incomplete week is excluded from the analysis, and all observations and source data are archived. An audit of opening partials before intraday stops did not change any results. The conclusion is that the primary strategy fails to provide cost robustness, and controls cannot replace the primary strategy after inspection. Adjusted dividend mechanics, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not related to Nagel's 48 industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a test. The test uses nine sector ETFs, which are a type of investment fund that tracks a specific sector, and uses daily bars, which are a type of data that shows the high, low, open, and close prices of a security over a given period, that have been adjusted for splits and dividends. The test ran from January 4, 2021, at 5:00:00 UTC, through September 10, 2026, at 4:00:00 UTC. The test buys the sector with the worst return over the prior five sessions, but only if that return is negative, and it does so at the next open, filling the order at the prior close price, using a cash-only approach that allows for fractional, or partial, shares. The test also has a time-only weekly exit control, which means that positions are closed based on time, not on price movements. The test includes fees of 1 basis point, which is a unit of measurement for interest rates and fees, plus slippage of 15 basis points on each side of the trade, which means that the test accounts for the difference between the expected price of a trade and the actual price. The test uses a vectorbt, which is a type of backtesting framework, long-only approach, which means that it only buys securities, and it reconciles, or matches, the daily equity, which is the total value of the portfolio, every day. The test also uses a cash-sharing approach, which means that it allocates cash to different parts of the portfolio. The mean trade confidence interval, which is a statistical measure of the range of possible values for a trade, is between -0.42498062146571663% and 0.2751204592455909% at a 95% confidence level, which is a statistical measure of the reliability of the results, based on 5000 bootstrap samples, which is a type of statistical simulation. The reasons for the test include a time value of 255, and the test had no partial exits, which are exits from a position that are not complete, and no same-day final exits, which are exits from a position that occur on the same day as the entry. The current incomplete week was excluded from the test, and all observations and source data were archived, and an audit of the opening partial before intraday stop, which is a type of risk management strategy, did not change any of the results. The conclusion of the test is that the primary strategy fails to meet the cost robustness, which means that it is not profitable after accounting for costs, and that the controls, which are the parts of the test that are designed to manage risk, cannot replace the primary strategy after holdout inspection, which is a type of statistical analysis. The test also notes that several mechanics, including adjusted dividend mechanics, actual 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and PDT, which is a type of regulatory rule, remain unvalidated, which means that they have not been tested or proven. The source of the data for the test is https://runway.adityaperswal.workers.dev/data/stock-bars, and this adaptation is not related to Nagel's 48 industry daily short portfolio, which is a type of investment strategy.
A frozen note, number 2174, explains the premises of a trading strategy. The strategy uses nine sector ETFs, which are a type of investment fund that tracks a specific sector, and daily bars, which are a type of data that shows the high, low, open, and close prices of a security over a given period, adjusted for splits and dividends, from January 4, 2016, to December 31, 2020. The strategy buys the sector with the worst return over the prior five sessions if it is negative, using the prior close as a signal to fill the next open, and only uses cash, with fractional shares allowed. The strategy exits at the end of each week, based on time only. The fees are one basis point, which is a unit of measurement for interest rates and fees, plus 15 basis points for slippage, which is the difference between the expected price and the actual price, on each side of the trade. The strategy uses a vectorized backtesting framework, which is a way of testing a strategy using historical data, with four phases per day, and is reconciled daily. The mean trade confidence interval, which is a statistical measure of the range of possible values, is between -0.46% and 0.14% at a 95% confidence level, based on 5000 bootstrap samples, which is a statistical method for estimating the variability of a sample. The reasons for the strategy include a time parameter of 220. The strategy does not use partial fills or same-day final exits. The current incomplete week is excluded from the analysis, and all observations and source data are archived. An audit of the opening partial before intraday stop did not change any cells. The conclusion is that the primary strategy fails to be robust, and the controls cannot replace the primary strategy after inspecting the holdout data, which is a portion of the data set aside for testing. The mechanics of adjusted dividends, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not related to Nagel's 48-industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a test. The test uses nine sector ETFs, with daily bars adjusted for dividends and splits, from January 4, 2021, 05:00:00+00:00 to September 10, 2026, 04:00:00+00:00. The test buys the sector with the worst return over the prior five sessions, but only if that return is negative, and it uses the prior close as a signal to fill the next open, with one cash-only fractional sleeve. The test exits weekly based on time only. It has fees of 1 basis point plus 5 basis points for slippage on each side, with four phases per day, using vectorbt for long-only cash sharing, reconciled daily, with no hidden leverage. The mean trade confidence interval at 95% is between -0.2173690604475051% and 0.48858693110590634%, based on 5000 bootstrap samples clustered by month. The reasons for this include a time factor of 255, with no partials and no same-day final exits. The current incomplete week is excluded, and all observations and sources are archived, with an audit of opening partials before intraday stops that did not change any cells. The conclusion, for research only, is that the primary test fails to be robust to costs, and controls cannot replace the primary test after inspecting the holdout. Adjusted dividend mechanics, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not Nagel's 48-industry daily short portfolio.
A frozen note, number 2174, explains the premises of a trading strategy. The strategy uses nine sector ETFs, with daily bars adjusted for dividends and splits, from January 4, 2016, to December 31, 2020. The strategy buys the sector with the worst return over the prior five sessions, but only if that return is negative, and it does so at the next open, using a signal from the prior close. The strategy uses a cash-only approach, with fractional shares allowed, and it exits positions based on time only, once a week. The strategy also includes fees of 1 basis point, plus a 5 basis point slippage on each side of the trade. The strategy has four phases per day and uses a vectorized backtesting approach, with long-only positions and cash sharing, reconciled daily. The strategy does not use hidden leverage. The mean trade confidence interval is between -0.26% and 0.33%, based on 5,000 bootstrap samples, clustered by month. The reasons for the strategy include a time factor, with 220 time periods considered. The strategy excludes the current incomplete week and archives all observations and sources. An audit of the opening partial and intraday stop did not change any results. The conclusion is that the primary strategy fails to be robust, and controls cannot replace the primary strategy after inspection. The strategy still needs to validate its mechanics, including adjusted dividend handling, 15-minute monitoring, and slippage estimates. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This strategy is not related to Nagel's 48-industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a test. The test uses nine sector ETFs, with daily bars that are adjusted for dividends and splits, from January 4, 2021, at 05:00:00+00:00 to September 10, 2026, at 04:00:00+00:00. The first session of each week buys the sector that had the worst return over the prior five sessions, but only if that return was negative. The purchase is made at the next open, with a signal from the prior close, and is a cash-only transaction that allows for fractional shares. A stop loss is set at one times the average true range over 20 days, with a target of three times that range, and an additional two times that range for a quarter, and a breakeven point the following day. The test also includes a gap open, conservative barriers, and a weekly time exit. Fees are one basis point plus 15 basis points for slippage on each side of the trade. The test has four phases per day, using a vectorized backtesting approach, and is long-only, with cash sharing reconciled daily, and no hidden leverage. The mean trade confidence interval at 95% is between -0.31652818404002264% and 0.2633172689313565%, based on 5000 bootstrap samples clustered by month. The reasons for certain decisions are given, with a gap target of 1, a stop of 106, a time of 123, a gap stop of 17, and a target of 8. There were 35 partial trades and 30 same-day final exits. The current incomplete week is excluded from the test, and all observations and source data are archived. An audit of the opening partial before the intraday stop did not change any results. The conclusion of the test is that the primary strategy fails the cost robustness test, and controls cannot replace the primary strategy after inspecting the holdout data. The test did not validate certain mechanics, including adjusted dividend mechanics, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This test is not related to Nagel's 48-industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a trading strategy. The strategy involves nine sector ETFs, with daily bars adjusted for dividends and splits, from January 4, 2016, to December 31, 2020. Each week, the strategy buys the sector with the worst return over the previous five sessions, but only if that return is negative. The buy signal is based on the prior day's close and the next day's open, with a cash-only fractionalsleeve, which means a portion of the investment is in cash. The strategy uses a stop-loss, or a price level at which to automatically sell, based on the average true range, or ATR, a measure of price volatility, over 20 days, with a target of three times the risk, or R, and a barrier, or price level, two times the R above the target. The strategy also has a weekly time exit, meaning it sells at the end of each week. There are fees of one basis point, or 1bp, plus slippage, or the difference between the expected and actual trade price, of 15bp on each side of the trade. The strategy has four phases per day and uses a vectorbt, a type of backtesting framework, for long-only, or buy-only, trades, with cash sharing reconciled every day. The strategy does not use hidden leverage, or borrowing to increase investment. The mean trade confidence interval, or CI, is between -0.45947011023146633 and 0.0412389443528169 percent, based on 5000 bootstrap samples, which are random samples of the data. The reasons for trades are: stop-loss 102 times, time exit 101 times, gap stop, or a stop-loss based on a large price gap, 11 times, and target 6 times. There were 22 partial trades and 32 same-day final exits. The current incomplete week is excluded from the analysis, and all observations and source data are archived. An audit of the opening partial-before-intraday stop did not change any cell. The conclusion is that the primary strategy fails to be cost-robust, and controls cannot replace the primary strategy after inspection. Adjusted dividend mechanics, actual 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader, or PDT, rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not based on Nagel's 48-industry daily short portfolio.
A news scan found the note: Frozen note2174. The premises of this test are: 1. Nine sector ETFs, with all-adjusted daily bars, from 2021-01-04 05:00:00+00:00 to 2026-09-10 04:00:00+00:00. The strategy buys the worst-performing sector from the previous 5 sessions if it is negative, using the prior close as a signal to fill at the next open, with one cash-only fractional sleeve. The strategy uses an ATR20 stop, which is a volatility-based stop-loss, with a target of 3R, and a break-even point the following day. It also uses a gap open, conservative barriers, and a weekly time exit. 2. The fees are 1 basis point plus slippage of 5 basis points on each side. The strategy has four phases per day, using a vectorbt long-only cash-sharing approach, reconciled every day, with no hidden leverage. 3. The mean trade confidence interval is between -0.14919233063956544% and 0.4504968275114483%, based on a month cluster of 5000 bootstrap samples. The reasons for the strategy's decisions are: gap target of 2, stop of 103, time of 128, gap stop of 15, and target of 7. There were 36 partial exits and 27 same-day final exits. 4. The current incomplete week is excluded, and all observations and sources are archived. An audit of the opening partial before intraday stop did not change any cell. The conclusion of this research is that the primary strategy fails the cost robustness test, and controls cannot replace the primary strategy after holdout inspection. The mechanics of adjusted dividends, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and PDT remain unvalidated. The source of this data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not Nagel's 48 industry daily short portfolio.
A frozen note, number 2174, outlines the premises of a trading strategy. The strategy involves nine sector ETFs, with daily price adjustments, from January 4, 2016, to December 31, 2020. Each week, the strategy buys the sector that performed the worst over the previous five sessions, but only if that performance was negative. The purchase is made at the next available opening price, with a fraction of the available cash. The strategy uses a stop-loss order, which is a type of order that automatically sells a security when it falls to a certain price, in this case, 1 times the average true range over 20 days, and a target price, which is 3 times the average true range. If the target is reached, the strategy aims to make an additional 2 times the average true range. The strategy also includes a time exit, which means that the position is closed at the end of the week. There are fees of 1 basis point, plus a slippage of 5 basis points on each side of the trade. The strategy is long-only, meaning it only buys securities, and it shares cash among the different positions. The strategy does not use leverage, which means it does not borrow money to make trades. The mean trade confidence interval is between -0.21% and 0.28%, based on 5000 bootstrap samples. The reasons for closing trades are: gap stop, which means the price gap at the opening, 12 times, time exit, 114 times, stop-loss order, 88 times, and target price, 6 times. There were 26 partial exits and 25 same-day final exits. The current incomplete week is excluded from the analysis, and all observations and source data are archived. An audit of the opening partial before intraday stop did not change any of the results. The conclusion is that the primary strategy fails to be robust, and the controls cannot replace the primary strategy after inspecting the holdout data. The mechanics of adjusted dividends, 15-minute monitor fills, fractional to whole-share partial exits, slippage estimates, and pattern day trader rules remain unvalidated. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. This adaptation is not related to Nagel's 48 industry daily short portfolio.
This report is based on premises that include evidence for momentum from a 2014 study by AQR, which is a type of investment management firm. The study, found at https://www.aqr.com/-/media/AQR/Documents/Journal-Articles/JPM-Fact-Fiction-and-Momentum-Investing.pdf, is about fact and fiction in momentum investing. The report also references a frozen note from 2024 about nine original sector ETFs, which are a type of investment fund. The note explains that the prior-month-end close price divided by the close price 189 days earlier, minus 1, is used to rank the top 3 sectors. The next session's open price is then used, with three fixed independent cash sleeves and no leverage, which means borrowing money to invest. The report also mentions a drop-rank monthly or 63-session exit, which is a strategy for selling investments, and a 2ATR stop and 6ATR target, where ATR stands for Average True Range, a measure of price volatility. Additionally, a 2R daily high makes the break-even point effective the following day. A gap open is used as an actual proxy, with the same bar stop first. The report also references data from https://runway.adityaperswal.workers.dev/data/stock-bars, which was requested from 2009 but only received from 2016 onward. The warm-up period means the first 252 rule trades from 2017, with no pre-2016 data. The frozen 2019 split was retained and not moved to improve results. The vectorbt long-only cash-sharing strategy is used, with every daily equity and trade count reconciled. A fee of 1 basis point each side, plus a 15 basis point slip, is applied. Corporate action total return is not independently rebuilt, and no taxes, cash yield, or operator costs are included. The results show 190 trades, with a mean return of 0.625286%, and an entry month cluster of 3000 bootstrap CI, which is a measure of confidence interval, of [-0.437693, 1.761003]%. The p-value is 35.2632%, the W-value is 8.1038%, and the L-value is 3.4484%. The plug-in half Kelly risk is 0.038580. The primary 252 training was negative, but the positive 189 neighbor cannot replace the pre-registered rule. The historical holdout may overlap prior research, and the daily break-even proxy is not 15-minute live proof. The report rejects live adoption and retains the positive holdout for prospective research. The artifacts are found in research/test_sector.py, sector_results.json, and percelltrades/returns.
This report is based on two premises. The first premise is that a 2014 study by AQR, which ranked stocks by momentum over a 12-month period, skipping the most recent month, provides evidence for momentum investing, but this specific strategy does not follow that approach. The study can be found at https://www.aqr.com/-/media/AQR/Documents/Journal-Articles/JPM-Fact-Fiction-and-Momentum-Investing.pdf. The second premise, noted in Frozen note 2124, involves nine original sector ETFs, where the momentum is calculated as the difference between the closing price 21 days ago and 189 days ago, and the top three positive sectors are selected. The next session's open is used, with three fixed independent cash sleeves and no leverage. The strategy exits positions monthly or after 63 sessions, using a stop-loss of 6 ATR and a target of 2 ATR. If the daily high is 2R above the entry price, the stop-loss becomes effective the following day. A gap open is used as a proxy for the actual open, with the same bar stop as the first stop. The data used was requested from 2009, but only data from 2016 onwards was received. The warm-up period means that the first 252 trades in 2017 were used to initialize the strategy, with no data used before 2016. The 2019 split was retained and not moved to improve results. The strategy uses vectorbt for long-only cash sharing, with every daily equity and trade count reconciled. The fees are 1 basis point for each side, plus a 15 basis point slip. Corporate actions are included in the total return, but not independently rebuilt, and no taxes, cash yield, or operator costs are included. The strategy made 55 trades, with a mean return of 0.497706%, and an entry month cluster of 3000 bootstrap CI of [-1.106902, 2.078437]%. The plug-in half Kelly risk is 0.042640. The primary 252 training period was negative, but the positive 189 neighbor cannot replace the preregistered rule. The historical holdout may overlap prior research, and the daily BE proxy is not a 15-minute live proof. The strategy is rejected for live adoption, but the positive holdout is retained for prospective research. The research artifacts can be found in research/test_sector.py, sector_results.json, and percelltrades/returns.
This report is based on premises that include evidence for momentum from a 2014 AQR study, which is a research paper on fact and fiction in momentum investing. The study can be found at https://www.aqr.com/-/media/AQR/Documents/Journal-Articles/JPM-Fact-Fiction-and-Momentum-Investing.pdf. The report also references a frozen note from 2024, which details a strategy using nine original sector ETFs, where the prior-month-end close is compared to the close 189 days prior, and the top three positive sectors are selected. The next session's open is used, with three fixed independent cash sleeves and no leverage. The strategy includes a monthly or 63-session exit, with a stop loss set at 6 ATR, which stands for Average True Range, a measure of volatility. The strategy also includes a target at 6 ATR and a daily high that makes the break-even point effective the following day. A gap open is used as an actual proxy, with the same bar stop as the first stop. The report uses data from 2016 onward, with a warm-up period that means the first 252 trades in 2017 are used, and there is no pre-2016 data. The frozen 2019 split is retained and not moved to improve results. The vectorbt, which is a backtesting library, is used for long-only cash sharing, with every daily equity and trade count reconciled. The fee is 1 basis point, or 0.01%, each side, plus a 5 basis point slip, which is a small cost. The corporate action total return is not independently rebuilt, and there are no taxes, cash yields, or operator costs. The results show 188 trades, with a mean return of 0.890330%, and an entry month cluster of 3000 bootstrap confidence intervals of [-0.188323, 2.061320]%. The p-value is 36.1702%, the W-value is 8.3338%, and the L-value is 3.3276%. The plug-in half Kelly risk is 0.053417, which is a measure of risk. The primary 252 training is negative, and the positive 189 neighbor cannot replace the pre-registered rule. The historical holdout may overlap prior research, and the daily break-even proxy is not a 15-minute live proof. The report rejects live adoption and retains the positive holdout for prospective research. The artifacts of the research can be found in the files research/test_sector.py, sector_results.json, and percelltrades/returns.
This report is based on two premises. The first premise is that a 2014 study by AQR, which ranked momentum based on a 12-month period with a 1-month skip, is evidence for momentum, but this adapter does not use that method. The study can be found at https://www.aqr.com/-/media/AQR/Documents/Journal-Articles/JPM-Fact-Fiction-and-Momentum-Investing.pdf. The second premise, from a frozen note numbered 2124, involves nine original sector ETFs and a specific trading strategy. The strategy involves calculating the momentum of each sector based on the prior month's end close and the close 189 days prior, then selecting the top three positive sectors. The next session's open is used to determine the trade, with three fixed independent cash sleeves and no leverage. A stop-loss is set at 6 ATR, which stands for Average True Range, a measure of volatility. The strategy also includes a target at 6 ATR and a daily high that makes the break-even point effective the following day. A gap open is used as an actual proxy, with the same bar stop first. The data used for this strategy was requested from 2009, but only data from 2016 onward was received. The warm-up period means that the first 252 trades in 2017 were used to establish the baseline, with no data used from before 2016. The frozen 2019 split was retained and not moved to improve results. The vectorbt long-only cash-sharing strategy was used, with every daily equity and trade count reconciled. A fee of 1 basis point, or 0.01%, was charged on each side, plus a 5 basis point slip. Corporate actions and total return were not independently rebuilt, and no taxes, cash yield, or operator costs were considered. The results show 54 trades, with a mean return of 0.689527%. The entry month cluster had a 95% confidence interval of -0.960180% to 2.353536%, with a p-value of 37.0370% and a W-value of 5.9756%. The plug-in half Kelly risk was 0.057695. The primary 252 training period was negative, and the positive 189 neighbor cannot replace the pre-registered rule. The historical holdout may overlap prior research, and the daily break-even proxy was not live-proofed at 15 minutes. As a result, live adoption was rejected, but the positive holdout was retained for prospective research. The artifacts from this research can be found in the files research/test_sector.py, sector_results.json, and percelltrades/returns.
This report is based on premises that include evidence for momentum from a 2014 study by AQR, which can be found at https://www.aqr.com/-/media/AQR/Documents/Journal-Articles/JPM-Fact-Fiction-and-Momentum-Investing.pdf. The report also references a frozen note, number 2124, which describes a strategy using nine original sector ETFs. The strategy involves ranking these ETFs based on their performance over the past 252 days, excluding the most recent month, and then selecting the top three. The next session's open is used to determine the entry point, with three fixed independent cash sleeves and no leverage. The strategy also includes a stop-loss and target, with a daily high of 2R making the stop-loss effective the following day. A gap open is used as a proxy, with the same bar stop being the first stop. The data used for this report was requested from 2009, but only data from 2016 onward was received. The warm-up period means that the first 252 rule trades were in 2017, with no pre-2016 data used. The frozen 2019 split was retained and not moved to improve results. The vectorbt long-only cash-sharing strategy was used, with every daily equity and trade count reconciled. The fees were 1 basis point on each side, plus a 15 basis point slip. The corporate action total return was not independently rebuilt, and no taxes, cash yield, or operator costs were included. The results show 190 trades, with a mean return of 1.011710%, and an entry month cluster of 3000 bootstrap CI of [-0.048485, 2.130268]%. The p-value was 38.4211%, the W-value was 8.2380%, and the L-value was 3.4970%. The plug-in half Kelly risk was 0.061405. The primary 252-day training period showed negative results, but a positive 189-neighbor cannot replace the pre-registered rule. The historical holdout may overlap prior research, and the daily BE proxy is not a 15-minute live proof. As a result, live adoption is rejected, but the positive holdout is retained for prospective research. The artifacts from this research can be found in the files research/test_sector.py, sector_results.json, and percelltrades/returns.