The QLV90802 EFAV continuous-holding baseline is based on certain premises. Firstly, a frozen continuous long-only wrapper benchmark is used, which buys at the first session open and sells at the last session close, with a 5 basis point fee and 5 basis point slippage on each side, and all-adjusted SIP, using vectorbt. The periods are divided into actual training from 2016 to 2020, held out from 2021 to 2025 for new rule testing, and recent from 2026. The results show a training return of 33.159539%, a drawdown of 27.477904%, an annual volatility of 14.410530%, and 1259 bars; a holdout return of 33.799286%, a drawdown of 27.455552%, an annual volatility of 11.609404%, and 1255 bars; and a recent return of 11.126209%, a drawdown of 6.658791%, an annual volatility of 11.623276%, and 170 bars. Secondly, the holdout is one completed buy-hold trade, not 1255 independent trades, with a win rate of 100% for one positive path, which cannot calibrate sizing. The full notional results are diagnostic, not a proposed allocation. Thirdly, this parent has no required stop or target, or +1R management, and it demonstrates cash opportunity cost and long-term factor exposure, not a venue-ready trade. Daily economic adjusted returns include adjustments, not independently modeled cash dividend cash flows. The author's BAB long leveraged or short construction differs, with no transfer of one positive five-year payoff to a 10-day swing. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the paper in analysis is 830. Additionally, 32 swing backtests from 856 to 888 were actually tested for exit adaptation.
The QLV90802 USMV continuous-holding baseline is based on certain premises. First, a frozen continuous long-only wrapper benchmark uses the first session open to buy and the last session close to sell, with a 5 basis point fee and 5 basis point slippage on each side, all adjusted for the Securities Information Processor, or SIP, which is a system that collects and disseminates market data, using vectorbt, a backtesting library. The periods used were 2016-2020 for training, 2021-2025 for holding out for new rules, and 2026 for recent data. The results for the training period were a return of 81.409784%, a drawdown of 33.071749%, an annual volatility of 16.029441%, and 1259 bars, or units of time. The holdout period results were a return of 50.061488%, a drawdown of 17.923149%, an annual volatility of 12.551699%, and 1255 bars. The recent period results were a return of 7.431228%, a drawdown of 6.456596%, an annual volatility of 8.873244%, and 170 bars. Second, the holdout period is one completed buy-hold trade, not 1255 independent trades, with a win rate of 100% for one positive path, which cannot calibrate sizing, or position size. The full notional results are diagnostic, not a proposed allocation. Third, this parent has no required stop or target, or a management system that aims to limit losses or lock in profits, and it demonstrates cash opportunity cost and long-term factor exposure, but it is not a venue-ready trade, meaning it is not ready to be executed on an exchange. Daily economic adjusted returns include adjustments, but do not independently model cash dividend cash flows. Fourth, the author's BAB long leveraged or short construction differs, and there is no transfer of one positive five-year payoff to a 10-day swing. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the paper in analysis is 830. Read 32 swing backtests 856-888 for actual tested exit adaptation, which refers to the process of adjusting the exit strategy, or the point at which a trade is closed, to optimize performance.
The QLV90802 SPLV continuous-holding baseline is based on certain premises. Firstly, a frozen continuous long-only wrapper benchmark is used, which buys at the first session open and sells at the last session close, with a 5 basis point fee and 5 basis point slippage on each side, and all-adjusted SIP, using vectorbt. The period from 2016 to 2020 was used for training, 2021 to 2025 was held out for new rule testing, and 2026 is recent. The results for the training period show a return of 63.735215%, a drawdown of 36.259259%, an annual volatility of 17.260015%, and 1259 bars. For the holdout period, the return was 40.318483%, the drawdown was 17.265390%, the annual volatility was 12.550622%, and there were 1255 bars. Recently, the return was 6.073193%, the drawdown was 7.416112%, the annual volatility was 10.941085%, and there were 170 bars. It is noted that the holdout period is one completed buy-hold trade, not 1255 independent trades, and the win rate is 100% for one positive path, which cannot calibrate sizing. The full notional results are diagnostic and not proposed allocation. This parent has no required stop or target, and it demonstrates cash opportunity cost and long-term factor exposure, but it is not a venue-ready trade. Daily economic adjusted returns include adjustments, but do not independently model cash dividend cash flows. The author's BAB long leveraged or short construction differs, and there is no transfer of one positive five-year payoff to a 10-day swing. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the paper in analysis is 830. Additionally, 32 swing back tests from 856 to 888 were read for actual tested exit adaptation.
A frozen note, number 877, describes a trading strategy. The strategy was trained on data from 2016 to 2020, and then tested on new data from 2021 to 2025. The test period was also historically inspected. The strategy only considers complete trading periods, which are called cohorts, from 2026. The strategy buys when the first session open of the month is higher than the previous month's close and also higher than the 100-day moving average, which is a calculation of the average price over the last 100 days. The previous bar's average true range, or ATR, which is a measure of price volatility, is used to set a stop loss, which is an order to sell when the price falls to a certain level, and a target price, which is the desired sale price. The strategy sells half of the position when the price reaches one average true range above the buy price, and then moves the stop loss to the break-even point, which is the price at which the trade will not result in a loss. If the price moves in the opposite direction, the strategy sells the remaining half of the position. The strategy also uses a gap stop, which is a stop loss that is triggered when the price gaps, or jumps, above or below a certain level. All prices are adjusted for dividends, which are payments made to shareholders, and a 5 basis point fee, which is a small charge, and 20 basis points of slippage, which is the difference between the expected and actual prices of a trade. The strategy uses a diagnostic tool to test the effect of different position sizes, starting with an initial cash amount of 100. The strategy does not use leverage, which is borrowing money to invest. The results of the strategy are compared to a benchmark using a tool called vectorbt, which reconstructs the trades and agrees with the results to within a very small margin of 3.55e-13 percentage points. The drawdown, which is the peak-to-trough decline in the value of the portfolio, is calculated using synthetic daily fill and close prices, rather than actual 15-minute prices. The positions are normalized to full allocation, which means that the entire portfolio is invested, for comparison purposes. A bootstrap analysis, which is a statistical test, was run 10,000 times with a seed value of 90802, and the results showed a 95% mean interval of -0.768018% to -0.193765%. The strategy only uses historical data to size positions, and any failures during the training or holdout periods are kept at zero, even if the actual result was positive, as in the case of the 2026 counterexample. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The strategy is not an exact leveraged betting against beta strategy. The study has 252 metrics observations for QLV90802, and is a deterministic study, which means that the results are not random, and is run using a local Python script called deterministicstudy.py.
A frozen note, number 877, was used for analysis. The analysis used data from 2016 to 2020 for training, and data from 2021 to 2025 was used as a holdout period, but was also historically inspected. The analysis only included completed cohorts from 2026. The strategy used was to buy at the first session open of the month if it was higher than the 100-day moving average of the previous month's close. The stop loss was set at 1 average true range, and the target was set at 1.5 average true ranges, with a time exit at the 10th session close. A moving average is a way to smooth out price data. The average true range is a measure of price volatility. The analysis did not include partial profits, and the gap stop open was used if there was ambiguity. All results were adjusted for daily systematic investment plan, with a 5 basis point fee and 20 basis points of slippage for each fill. A basis point is one hundredth of a percent. The analysis used fractional continuous sizing, with an initial cash amount of 100, one position, and no leverage. Leverage refers to using borrowed money to invest. The vectorbt exact fill reconstruction agreed with independent compounding within 3.80e-13 percentage points. The drawdown was calculated using synthetic per day fill and close slots, not the actual 15-minute path. Positions were normalized for full allocation for comparison, but not as a proposal. A bootstrap analysis was done with 10,000 iterations and a seed of 90802, resulting in a 95% mean interval of -0.879100% to -0.205256%. The historical profit and loss sizing was only used for training and holdout, and failures were kept at actual 0 even if there were positive counterexamples in 2026. The source of the data was https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper was https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The analysis was not an exact leveraged betting against beta strategy. The full 252 metrics observations for QLV90802 were used, and the study was deterministic, using the ledgers local in study.py.
A frozen note, number 877, contains information about a trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is not available. The years 2021 to 2025 were used as a holdout period, but the market during this time was also historically inspected. The year 2026 was used to complete cohorts. The strategy buys at the open of the first session of the month if the previous month's close is greater than the 100-day moving average, which is a calculation of the average price of a stock over 100 days, used to determine trends. The strategy also uses a stop-loss, which is a price level at which to sell a stock to limit losses, set at one average true range, a measure of volatility, below the buy price. The target price is set at 1.5 average true ranges above the buy price, and the position is closed at the close of the 10th session. If the stock price reaches one average true range above the buy price, half of the position is sold, and the stop-loss is moved to the break-even price. The strategy also uses a gap stop, which is a stop-loss set at the open price if the stock gaps up or down, and the target price is capped. All prices are adjusted for dividends and splits, and a fee of 5 basis points, or 0.05%, and slippage of 5 basis points are applied to each fill, which is the execution of a trade. The strategy uses fractional continuous sizing, which means that the position size is adjusted based on the stock's volatility, and the initial cash is $100. The strategy does not use leverage, which is borrowing money to buy stocks. The vectorbt exact fill reconstruction agrees with independent compounding within 1.01e-13 percentage points, which means that the strategy's performance is accurate to a very small degree. The drawdown, which is the peak-to-trough decline in the stock's price, uses synthetic per-day fill/close slots, not the actual 15-minute path, and positions are normalized to full allocation for comparison, not as a proposal. A bootstrap analysis, which is a statistical method for estimating the uncertainty of a strategy's performance, was performed with 10,000 iterations and a seed of 90802, and the 95% mean interval is [-0.840583, 1.131843]%. The historical profit/loss and win/loss sizing are only used for training, and failures in training and holdout keep the actual profit/loss at zero, even if it is positive in 2026. The strategy's performance is compared to the EEMV, which is an exchange-traded fund that tracks the performance of emerging markets. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The strategy is not an exact leveraged betting against beta strategy. The QLV90802 metrics have 252 observations, and the study is deterministic, meaning that it is based on historical data and does not involve random chance. The ledgers are local and can be found in the deterministic study.py file.
A frozen note, number 877, describes a trading strategy. The strategy was trained on market data from 2016 to 2020, and tested on data from 2021 to 2025. It was also applied to historical market data, and the results for completed trades in 2026 were analyzed. The strategy buys when the first session open of the month is higher than the previous month's close and the 100-day moving average. It sets a stop loss at 1 ATR, or average true range, and a target at 1.5 ATR, with a time exit at the 10th session close. If the trade reaches 1 ATR in profit, it sells half and moves the stop to breakeven. The strategy also uses a gap stop and adjusts for slippage and fees. It uses a fractional continuous sizing diagnostic and starts with 100 units of cash. The strategy does not use leverage and has a single position. The results were verified using vectorbt exact fill reconstruction, which agreed with independent compounding within 3.92e-13 percentage points. The drawdown was calculated using synthetic per-day fill/close slots. A bootstrap analysis with 10,000 iterations and a seed of 90802 found a 95% mean interval of -0.469122% to 0.117191%. The strategy's performance was compared to historical data, and the results were used to size trades. The source of the market data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the strategy is related to a paper by Frazzini and Pedersen, available at https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The strategy is not an exact leveraged betting against beta strategy. The full metrics for QLV90802 are available, with 252 observations, and the study was deterministic, using a local ledger.
A frozen note, number 877, describes the training of an automated trading system. The system was trained on market data from 2016 to 2020, and data from 2021 to 2025 was used to test the system with new rules, but the market period was also inspected historically. The system was fully trained by 2026, using only completed cohorts. The system buys when the monthly session open is higher than the previous month's close and also higher than the 100-day moving average. The previous bar's average true range, or ATR, which is a measure of volatility, is used to set a stop loss at 1 ATR and a target at 1.5 ATR, with a time exit at the 10th session close. When the system reaches a profit of 1 ATR, it sells half of the position and moves the stop loss to break even, meaning it will not lose money if the market moves against it. If the market moves against the system, it will sell the rest of the position at the same bar, which is a unit of time. The system also uses a gap stop, which means it will stop trading if there is a large gap in the market price. All adjustments are made daily, and there is a fee of 5 basis points and a slippage of 5 basis points for each trade. The system uses fractional continuous sizing, which means it adjusts the size of the trade based on the available cash, and it starts with an initial cash amount of 100. The system does not use leverage, which means it does not borrow money to make trades. The vectorbt exact fill reconstruction agrees with independent compounding within 3.55e-13 percentage points, which is a measure of the system's accuracy. The drawdown, which is the measure of the system's losses, uses synthetic per day fill and close slots, not the actual 15-minute path. Positions are normalized to full allocation for comparison, but not as a proposal. The system uses a bootstrap, which is a statistical method, with 10000 seeds and a 95% mean interval of [-0.566171, 0.097840]%. The historical profit and loss sizing is only used for training, and failures in training and holdout keep the actual profit at 0, even if there are positive counterexamples in 2026. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The system is not an exact leveraged betting against beta system. The system has 252 metrics observations for QLV90802 metrics, and it uses a deterministic study with a local ledger.
A frozen note, number 877, contains information about the QLV90802 EEMV trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is unavailable. The years 2021 to 2025 were used as a holdout period, but the market during this time was also historically inspected. The year 2026 was used to complete cohorts. The strategy buys on the first session open of each month if the previous month's close is greater than the 100-day moving average, which is a calculation of the average price of a stock over the last 100 days. The previous bar's average true range, or ATR, which measures the volatility of a stock, is used to set a stop loss, which is an order to sell a stock when it falls to a certain price, at one ATR, and a target, which is the expected price of a stock, at 1.5 ATR. The strategy also has a time exit, which means it sells the stock at the close of the 10th session. The strategy does not allow partial profit, which means it does not sell a portion of the stock to lock in profit. A gap stop open, which is a stop loss that is triggered when a stock opens at a certain price, is used, and the stop loss is triggered first if it is ambiguous. The target is capped, which means it has a maximum value. All adjustments are made using the Security Information Processor, or SIP, daily, and a fee of 5 basis points, or 0.05%, and a slippage, which is the difference between the expected price of a stock and the actual price, of 5 basis points are applied to each fill, which is an order that is executed. The strategy uses fractional continuous sizing, which means it adjusts the size of the position based on the available cash, and it starts with an initial cash amount of 100. The strategy has one position and does not use leverage, which is borrowing money to buy more stock. The vectorbt exact fill reconstruction agrees with independent compounding within 6.53e-14 percentage points, which means the strategy's returns are accurate to a very small degree. The drawdown, which is the decrease in value of a stock, uses synthetic per day fill and close slots, not the actual 15-minute path, and positions are normalized to full allocation for comparison, not as a proposal. A bootstrap, which is a statistical method, of 10,000 trades with a seed of 90802 was used to estimate the mean interval, which is the range of possible returns, and it was found to be between -1.591759% and 1.210684%. The historical profit and loss, or P/L, sizing is only used for training and holdout, and failures in training and holdout keep the actual value at 0, even if it is positive in 2026. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The strategy is not an exact leveraged betting against beta, or BAB, strategy. The study has 252 metrics observations for QLV90802 metrics, and it is a deterministic study, which means it is based on known variables, and it uses the ledgers local in the study.py file.
A frozen note, number 877, contains information about a trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is not available. The strategy was tested using new rules from 2021 to 2025, which is a historically inspected market period, and completed cohorts were used in 2026. The strategy buys when the monthly first session open is greater than the previous month's close and also greater than the 100-day moving average, which is a calculation of the average price of a stock over 100 days, used to determine trends. The previous bar's average true range, a measure of volatility, over 14 days is used to set a stop loss, which is a price level at which to automatically sell a stock, and a target price, which is the expected price of a stock. The strategy uses a time exit, which means it sells the stock at the close of the 10th session. There is no partial profit, meaning the strategy does not sell a portion of the stock. A gap stop open, which is a stop loss that is triggered when the stock price gaps, or jumps, above or below a certain price, is used, and the stop loss is triggered first if it is ambiguous. The strategy also uses a target cap, which is a maximum price at which to sell the stock. All adjustments are made using the Security Information Processor, or SIP, daily, and a 5 basis point fee, which is a small percentage of the trade value, and 5 basis points of slippage, which is the difference between the expected price of a trade and the actual price, are added to each fill, which is the execution of a trade. The strategy uses fractional continuous sizing, which is a method of determining the size of a trade, and a diagnostic tool shows that the strategy would have started with 100 dollars of initial cash, and would have taken one position, with no leverage, which is the use of borrowed money to increase the size of a trade. The vectorbt exact fill reconstruction, a tool used to recreate the fills of a strategy, agrees with independent compounding, a method of calculating the returns of a strategy, within 2.43e-13 percentage points, which is a very small difference. The drawdown, which is the peak-to-trough decline in the value of a strategy, uses synthetic per day fill and close slots, rather than the actual 15-minute path, and positions are normalized to full allocation for comparison, but not as a proposal. A bootstrap, a statistical method used to estimate the variability of a strategy, was run 10,000 times with a seed of 90802, and the 95% mean interval was [-0.581388, 0.097071]%. The historical profit and loss sizing is only used for this strategy, and failures in the training and holdout periods keep the actual value at 0, even if there were positive counterexamples in 2026. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. This strategy is not an exact leveraged Betting Against Beta, or BAB, strategy. The study has 252 metrics observations for QLV90802 metrics, and is a deterministic study, which means that the results are not random, and is run using the deterministic study.py and ledgers local tools.
A frozen note, number 877, describes the training of a trading system called QLV90802 EEMV. The system was trained using data from 2016 to 2020, and data from 2021 to 2025 was used to test the new rules, but the market period was also inspected historically. The system only considers completed cohorts from 2026. The system enters a trade on the first session open of each month if the previous month's close is above the 100-day moving average, which is a calculation of the average price of a stock over the last 100 days. The system uses a stop-loss, which is a price level at which to automatically sell a stock, set at one average true range, a measure of volatility, below the entry price, and a target price set at 1.5 average true ranges above the entry price. The system also has a time exit, which means it will automatically sell a stock at the close of the 10th session. The system does not allow partial profits, meaning it will not sell a portion of a stock position. The system also uses a gap stop, which means it will stop a trade if the stock price gaps, or jumps, above or below a certain price level. All calculations are adjusted for dividends and splits, and a 5 basis point fee and 5 basis point slippage, which is the difference between the expected price of a trade and the actual price, are applied to each fill, which is the execution of a trade. The system uses fractional continuous sizing, which means it will adjust the size of a trade based on the available cash, and starts with an initial cash balance of 100. The system does not use leverage, which means it will not borrow money to buy stocks. The vectorbt exact fill reconstruction agrees with independent compounding within 3.27e-13 percentage points, which means the system's calculations are accurate to a very high degree. The drawdown, which is the peak-to-trough decline in the value of a stock or portfolio, uses synthetic per-day fill/close slots, not the actual 15-minute path, and positions are normalized for full allocation for comparison, but not as a proposal. The system uses a bootstrap, which is a statistical method for estimating the variability of a system, with 10,000 iterations and a seed of 90802, and the 95% mean interval is [-0.647948, 0.107154]%. The system only uses historical data for sizing, and failures in training and holdout keep the actual value at 0, even if the counterexample in 2026 is positive. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The system is not an exact leveraged betting against beta, which is a strategy that involves betting against the beta, or volatility, of a stock. The system has 252 metrics observations, and the study is deterministic, meaning it is based on fixed and known variables, and the ledgers are local, meaning they are stored on a local computer.
A frozen note, number 877, was written. The note is about a trading strategy that was trained on data from 2016 to 2020, and tested on data from 2021 to 2025. The strategy was also tested on historical market data, and the results for 2026 are complete. The strategy buys a stock at the start of the month if the stock's price at the start of the month is higher than its average price over the last 100 days. The strategy also uses a stop-loss order, which is an order to sell a stock when it falls to a certain price, in this case, 1 ATR, or average true range, below the buy price. The strategy sells half of the stock when it rises by 1 ATR, and then moves the stop-loss order to the break-even point. If the stock's price moves in the opposite direction of the expected direction, the strategy sells the rest of the stock at the same price. The strategy also uses a gap stop, which is a stop-loss order that is triggered when a stock's price gaps up or down at the start of trading. The strategy's performance is adjusted for a 5 basis point fee and 20 basis points of slippage per trade. The strategy uses a fractional continuous sizing diagnostic, which means that it trades a fraction of the total available cash, and it starts with 100 units of cash. The strategy does not use leverage, which means that it does not borrow money to trade. The Vectorbt exact fill reconstruction method was used to test the strategy, and it agreed with the independent compounding method within 4.69e-13 percentage points. The drawdown, or the measure of the strategy's worst performance, uses synthetic per-day fill and close slots, rather than the actual 15-minute path. The positions are normalized to full allocation for comparison, but this is not a proposal. A bootstrap test was run 10,000 times with a seed of 90802, and the 95% mean interval was between -0.849032% and -0.327937%. The historical profit and loss sizing is only used for this study, and failures in training and holdout keep the actual profit and loss at zero, even if it was positive in 2026. The data for this study came from https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. This study is not an exact leveraged betting against beta strategy. The QLV90802 metrics have 252 full observations, and the study is deterministic, meaning that it does not use random numbers. The ledgers are local and can be found in the deterministic study.py file.
A frozen note, number 877, was used to analyze the performance of a trading strategy. The analysis used data from 2016 to 2020 for training, and from 2021 to 2025 as a holdout period, but also looked at the market from a historical perspective. The analysis only considered completed trades from 2026. The strategy was to buy when the first session open of the month was higher than the previous month's close and also higher than the 100-day moving average. The strategy used a stop loss of one average true range, a target of 1.5 average true ranges, and a time exit after 10 sessions. The strategy did not allow for partial profits. A gap stop was used at the open, and the target was capped. All calculations were adjusted for daily settlement and included a 5 basis point fee and 20 basis points of slippage for each fill. The analysis used a fractional continuous sizing diagnostic, starting with 100 units of cash, one position, and no leverage. The results were verified using vectorbt exact fill reconstruction, which showed independent compounding within 3.13e-13 percentage points. The drawdown was calculated using synthetic daily fill and close slots, not the actual 15-minute path. Positions were normalized for full allocation for comparison purposes only. A bootstrap analysis was done with 10,000 iterations, seed 90802, and showed a 95% mean interval of -0.851523% to -0.199385%. The analysis only used historical data for sizing, and failures in training and holdout periods were kept at zero, even if there were positive counterexamples in 2026. The data source was https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper was https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The strategy was not an exact leveraged beta strategy. The analysis included 252 metrics observations for QLV90802 and was a deterministic study using local ledgers.
A frozen note, number 877, describes a trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is not available. The strategy was tested using new rules from 2021 to 2025, and also examined using historical market data from 2026. The strategy buys when the first session open of the month is higher than the previous month's close and the 100-day moving average. The stop loss is set at 1 average true range, and the target is set at 1.5 average true ranges, with a time exit at the 10th session close. When the price reaches 1 average true range above the buy price, half of the position is sold and the stop loss is moved to break even. If the trade is not going well, the same rules apply. The strategy also uses a gap stop open and stops at the first sign of trouble if it is unclear what to do, and the target is capped. All calculations are adjusted for daily settlement and include a 5 basis point fee and 5 basis point slippage for each fill. The strategy uses a diagnostic with continuous sizing and starts with 100 units of cash, one position, and no leverage. The results are verified using vectorbt exact fill reconstruction, which agrees with independent compounding within 2.22e-15 percentage points. Drawdown is calculated using synthetic daily fill and close slots, not the actual 15-minute path, and positions are normalized for full allocation for comparison, not as a proposal. A bootstrap analysis with 10,000 iterations and a seed of 90802 shows a 95% mean interval of -1.705693% to 0.823750%. The strategy only uses historical data for position and win/loss sizing, and failures in training and holdout keep the actual value at 0, even if there are positive counterexamples in 2026. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. This strategy is not an exact leveraged betting against beta strategy. The QLV90802 metrics have 252 observations, and the study is deterministic and can be found in the local ledgers and study.py file.
A frozen note, number 877, contains information about a trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is not available. The strategy was tested using new rules from 2021 to 2025, which is a historically inspected market period, and completed cohorts only in 2026. The strategy buys when the monthly first session open is greater than the previous month close and the 100-day moving average, which is a calculation of the average price of a stock over 100 days, and sells when the previous bar's average true range, which is a measure of volatility, is greater than 1 ATR, with a target of 1.5 ATR, and a time exit of the 10th session close. If the strategy makes a profit of 1 ATR, it sells half and stops at breakeven, which means it sets the stop-loss at the purchase price, and if the trade is adverse, it uses the same bar ordering. The strategy also uses a gap stop open and stops at the first sign of ambiguity, with a capped target. All calculations are adjusted for daily settlement and include a 5 basis point fee and 5 basis point slippage for each fill, which is a measure of the difference between the expected price and the actual price. The strategy uses fractional continuous sizing, which means it adjusts the position size based on the available cash, with an initial cash amount of 100 and one position, and no leverage, which means it does not use borrowed money. The vectorbt exact fill reconstruction agrees with independent compounding within 4.74e-13 percentage points, which is a measure of the accuracy of the strategy. The drawdown, which is a measure of the decline in value, uses synthetic per day fill and close slots, not the actual 15-minute path, and positions are normalized for full allocation for comparison, not proposal. The strategy uses IID trade bootstrap with 10000 seed 90802, which is a method of testing the strategy, and has a 95% mean interval of [-0.554674, -0.034357]%. The historical profit and loss sizing is only used for training and holdout, and failures in training and holdout keep the actual value at 0, even if it is positive, as seen in the 2026 counterexample. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The strategy is not an exact leveraged betting against beta, which is a type of investment strategy. The study has 252 metrics observations for QLV90802 metrics, and is a deterministic study using the ledgers local in the study.py file.
A frozen note, number 877, describes the training of a trading system. The system was trained on data from 2016 to 2020, and data from 2021 to 2025 was used to test the new rules, but the market period was also inspected historically. The system only uses completed data up to 2026. The system buys when the monthly first session open is higher than the previous month's close and also higher than the 100-day moving average, which is a calculation of the average price over the last 100 days. The system also uses a stop, which is a price level at which the system automatically sells, set at 1 average true range, or ATR, which is a measure of price volatility, below the previous bar's close, and a target, or selling price, set at 1.5 ATR above the previous bar's close. The system sells half of the position when it reaches a profit of 1 ATR, and then moves the stop to the break-even point, which is the price at which the trade was entered. If the system is stopped out, or sells due to a stop, on the same bar, or time period, it uses the same ordering. The system also uses a gap stop, which is a stop set at the open price if there is a large gap, or price jump, and stops at the first price if the stop is ambiguous, and caps the target price. All prices are adjusted for dividends and splits, which are changes in the stock's price due to the company paying out dividends or splitting the stock. The system uses a 5 basis point fee, which is 0.05%, and 5 basis points of slippage, which is the difference between the expected and actual price of a trade, for each fill, or trade execution. The system uses fractional continuous sizing, which means it adjusts the size of the trade based on the available cash, and starts with 100 dollars of initial cash, and only takes one position at a time, with no leverage, or borrowing. The system's results were reconstructed using the vectorbt exact fill method, which agrees with independent compounding within 3.41e-13 percentage points, which is a very small difference. The system's drawdown, or largest loss, uses synthetic per day fill and close slots, rather than the actual 15-minute path, and positions are normalized to full allocation for comparison, but not as a proposal. The system was tested using a bootstrap, which is a method of testing a system by randomly sampling the data, with 10,000 iterations, and a seed of 90802, and the results showed a 95% mean interval of [-0.450050, 0.041509]%. The system only uses historical profit and loss sizing, and failures in training and holdout keep the actual value at 0, even if the result is positive, as in the 2026 counterexample. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The system is not an exact leveraged betting against beta, or BAB, strategy. The system has 252 metrics observations for QLV90802, and the study is deterministic, meaning it always produces the same results, and is run using the study.py and ledgers local files.
A frozen note, number 877, contains information about the QLV90802 EFAV trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is unavailable. The years 2021 to 2025 were used as a holdout period to test the new rules, but the market performance during this time was also historically inspected. The year 2026 was used to complete the cohorts. The strategy buys when the monthly first session open is greater than the previous month's close and the 100-day moving average. The previous bar's average true range, or ATR, is used to set a stop loss at 1 ATR and a target at 1.5 ATR, with a time exit at the 10th session close. The strategy does not allow partial profits and uses a gap stop open, stopping first if ambiguous, with a capped target. All adjustments are made using the Security Information Processor, or SIP, daily, with a 5 basis point fee and 5 basis point slippage for each fill. The strategy uses fractional continuous sizing diagnostic, starting with 100 initial cash, one position, and no leverage. The vectorbt exact fill reconstruction agrees with independent compounding within 8.44e-15 percentage points. Drawdown uses synthetic per day fill and close slots, not the actual 15-minute path, and positions are normalized for full allocation for comparison, not proposal. A bootstrap of 10,000 trades with a seed of 90802 was used, resulting in a 95% mean interval of -1.656603% to 1.016394%. The historical profit and loss sizing is only used for training and holdout, and failures during training and holdout keep the actual value at 0, even if there is a positive counterexample in 2026. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. This strategy is not an exact leveraged Betting Against Beta, or BAB. The study includes 252 metrics observations for QLV90802 metrics, and is a deterministic study using the ledgers local in study.py.
This is a frozen note about the QLV90802 EFAV holdout. The training period is from 2016 to 2020, and the new-rule holdout period is from 2021 to 2025, with a historically inspected market period. The year 2026 has completed cohorts only. The strategy is to buy when the monthly first session open is greater than the previous month's close and the 100-day moving average. The stop loss is set at 1 ATR, or average true range, and the target is 1.5 ATR. The time exit is set at the 10th session close. There is an unmanaged parent for swing exit comparison, with no partial profit taken. A gap stop open is used, and the stop is triggered first if there is ambiguity. The target is capped. All adjustments are made using SIP daily, with a 5 basis point fee and 5 basis point slippage for each fill. The sizing is diagnostic, with 100 initial cash, one position, and no leverage. The vectorbt exact fill reconstruction agrees with independent compounding within 4.83e-13 percentage points. The drawdown uses synthetic per day fill and close slots, not the actual 15-minute path. Positions are normalized for full allocation for comparison, not proposal. A bootstrap of 10,000 trades was done with a seed of 90802, resulting in a 95% mean interval of -0.551373% to 0.100114%. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. This is not an exact leveraged BAB. There are 252 metrics observations for QLV90802 metrics, and the study is deterministic, using ledgers local in the study.py file.
A frozen note, number 877, describes the training of a trading system. The system was trained on data from 2016 to 2020, and data from 2013 to 2015 was not available. The system was tested on data from 2021 to 2025 using new rules, and also on historical market data from 2026. The system enters a trade when the monthly session open is higher than the previous month's close and also higher than the 100-day moving average, which is a calculation of the average price of a stock over the past 100 days. The previous bar's average true range, or ATR, which measures the volatility of a stock, is used to set a stop loss, which is a price level at which the system will automatically sell a stock to limit losses, and a target price, which is the price at which the system will automatically sell a stock to take profits. The system uses a time exit, which means it will automatically sell a stock after a certain number of sessions, in this case 10. The system does not allow partial profits, which means it will not sell a portion of a stock position to take profits. The system also uses a gap stop, which is a type of stop loss that is triggered when a stock's price gaps, or jumps, above or below a certain price level. All adjustments are made using a daily systematic investment plan, or SIP, which is a method of investing a fixed amount of money at regular intervals. The system also includes a 5 basis point fee, which is a small percentage of the investment amount, and 5 basis points of slippage, which is the difference between the expected price of a trade and the actual price. The system uses fractional continuous sizing, which means it will invest a fraction of the total investment amount in each trade, and it starts with an initial cash amount of 100. The system does not use leverage, which means it will not borrow money to invest. The vectorbt exact fill reconstruction method, which is a way of simulating trades, agrees with the independent compounding method, which is a way of calculating returns, within 3.84e-13 percentage points. The drawdown, which is the peak-to-trough decline in the value of an investment, uses synthetic per-day fill/close slots, which are simulated trades, rather than actual 15-minute trades. The positions are normalized to full allocation, which means they are adjusted to reflect the maximum possible investment amount, for comparison purposes only. The system uses a bootstrap method, which is a way of simulating trades, with 10,000 seeds, and the 95% mean interval is [-0.507839, 0.083461]%. The historical profit/loss and win/loss sizing is only used for training, and failures in training and holdout keep the actual value at 0, even if the 2026 counterexample is positive. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The system is not an exact leveraged betting against beta, or BAB, strategy. The system has 252 metrics observations, and the deterministic study is available in the ledgers local file.
A frozen note, number 877, contains information about a trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is not available. The strategy was tested using new rules from 2021 to 2025, and also examined during a historical market period. The strategy only considers completed market periods up to 2026. The strategy buys when the first session open of the month is higher than the previous month's close and also higher than the 100-day moving average, which is a calculation of the average price of a stock over 100 days. The strategy uses a stop loss, which is a price level at which to automatically sell a stock, set at 1 average true range, or ATR, which is a measure of how much a stock's price has moved, away from the buy price. The strategy also uses a target price, which is 1.5 ATRs away from the buy price, and a time exit, which is when the strategy sells the stock after 10 sessions. If the stock price reaches 1 ATR away from the buy price, the strategy sells half of the stock and moves the stop loss to the break-even point, which is the price at which the trade will not result in a loss. The strategy also uses a gap stop, which is when the stock price opens at a much higher or lower price than the previous day's close, and stops the trade if this happens. All stock prices are adjusted for dividends and splits, which are changes to the stock's price or number of shares. The strategy uses a 5 basis point fee, which is 0.05%, and 20 basis points of slippage, which is 0.20%, for each trade. The strategy uses a diagnostic to test the effect of different position sizes, starting with 100 dollars of cash, and does not use leverage, which is borrowing money to buy more stocks. The strategy's results are compared to a reconstruction using the vectorbt library, which is a tool for backtesting trading strategies, and the results agree within 4.19e-13 percentage points, which is a very small difference. The strategy's drawdown, which is the largest loss from peak to trough, is calculated using synthetic daily fills and closes, not actual 15-minute price paths. The strategy's positions are normalized to full allocation, which means that the entire amount of cash is invested, for comparison purposes. The strategy uses a bootstrap, which is a statistical method, with 10,000 iterations and a seed of 90802, to estimate the mean interval, which is the range of possible mean returns, of -0.895933% to -0.350419%. The strategy only uses historical data for position sizing, and failures in the training and holdout periods are kept at zero, even if the strategy had positive returns in 2026. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The strategy is not an exact leveraged betting against beta strategy. The strategy has 252 metrics observations, and the study is deterministic, meaning that it always produces the same results, and is run using the study.py and ledgers local tools.
A frozen note, number 877, was found. This note is about a trading strategy that was trained on data from 2016 to 2020, and then tested on data from 2021 to 2025, which is called a holdout period. The strategy was also tested on historical market data, and the results for 2026 are complete. The strategy buys a stock when the stock's price at the start of the month is higher than its average price over the last 100 days. The strategy sells the stock when it reaches a target price, which is 1.5 times the stock's average true range over the last 14 days, or when 10 sessions have passed, whichever comes first. A true range is a measure of a stock's volatility, which is a measure of how much its price changes. The strategy also has a stop-loss, which means it sells the stock if it loses a certain amount of value. The results of this strategy are compared to a strategy that does not manage its trades. The strategy's performance is adjusted for fees and other costs, and it is tested with a fictional $100 initial investment, with no borrowing of money, which is called leverage. The strategy's performance is also tested with a method called bootstrapping, which involves randomly re-sampling the data 10,000 times, with a random seed of 90802. The results show that the strategy's performance is within a certain range, with a 95% confidence interval of -0.810655% to -0.145435%. The strategy's performance is also compared to a study by Frazzini and Pedersen, which is available at https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The data used for this strategy is available at https://runway.adityaperswal.workers.dev/data/stock-bars.
A frozen note, number 877, contains information about the QLV90802 USMV trading strategy. The strategy was trained on market data from 2016 to 2020, and tested on data from 2021 to 2025. The strategy also considered historical market data from 2013 to 2015, although this data was not used for training, and from 2026. The strategy buys a stock if the first session open of the month is greater than the 100-day moving average of the previous month's close. The strategy uses a stop loss of 1 average true range, or ATR, which is a measure of market volatility, and a target of 1.5 ATR. If the stock price reaches 1 ATR above the buy price, the strategy sells half of the position and moves the stop loss to the break-even point. If the stock price moves in the opposite direction, the strategy sells the entire position at the stop loss price. The strategy also uses a gap stop, which means that if a stock gaps open, the strategy will sell the position at the open price. All trades are adjusted for slippage, which is the difference between the expected price of a trade and the actual price, and a fee of 5 basis points, or 0.05%, plus 5 basis points of slippage per fill. The strategy uses fractional continuous sizing, which means that the position size is adjusted based on the cash available, and starts with an initial cash position of 100. The strategy does not use leverage, which means that it does not borrow money to buy stocks. The Vectorbt exact fill reconstruction agrees with independent compounding within 7.77e-14 percentage points, which means that the strategy's performance is consistent with the expected performance. The drawdown, which is the maximum loss of the strategy, uses synthetic per-day fill/close slots, rather than the actual 15-minute path, and positions are normalized to full allocation for comparison. The strategy was tested using a bootstrap method, which involves randomly sampling the data with replacement, 10,000 times, with a seed of 90802, and the 95% mean interval was [-0.834121, 0.226791]%. The strategy only uses historical data for sizing, and failures in training and holdout keep the actual value at 0, even if the counterexample is positive in 2026. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. The strategy is not an exact leveraged betting against beta strategy. The study has 252 metrics observations for QLV90802, and is a deterministic study, which means that it uses a fixed set of rules to make decisions, and the results are based on a local ledger.
A frozen note, number 877, contains information about a trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is not available. The strategy was tested using new rules from 2021 to 2025, and also examined during a historical market period. The testing was completed using cohorts from 2026. The strategy buys when the monthly opening price is higher than the previous month's closing price and also higher than the 100-day moving average. The previous bar's average true range, or ATR, is used to set a stop loss and a target price, which is 1.5 times the ATR. The strategy sells half of the position when it reaches a profit of 1 ATR, and then moves the stop loss to break even. If the trade is not going well, the strategy sells the rest of the position at the same bar. The strategy also uses a gap stop, which means it stops trading if the price gaps too much. All prices are adjusted for dividends and splits, and a fee of 5 basis points and a slippage of 5 basis points are applied to each trade. The strategy uses a fractional continuous sizing diagnostic, which means it adjusts the position size based on the available cash, starting with 100 dollars and one position, without using leverage. The vectorbt exact fill reconstruction agrees with independent compounding within 3.27e-13 percentage points. The drawdown is calculated using synthetic per-day fill and close slots, not the actual 15-minute path. Positions are normalized to full allocation for comparison, but this is not a proposal. A bootstrap analysis with 10,000 iterations and a seed of 90802 shows a 95% mean interval of -0.598364% to -0.048365%. The strategy only uses historical data for position sizing, and failures during training and holdout periods keep the actual position size at zero, even if the strategy would have made a profit in 2026. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. This strategy is not an exact leveraged betting against beta strategy. The QLV90802 metrics have 252 observations, and the study is deterministic, using a local ledger in the study.py file.
A frozen note, number 877, contains information about a training period from 2016 to 2020, with data from 2013 to 2015 being unavailable. The period from 2021 to 2025 was used as a holdout, but it was also historically inspected. The year 2026 was used to complete cohorts. The training rule was to buy at the monthly first session open if it was higher than the 100-day moving average, which is a calculation of the average price of a stock over a 100-day period. The previous bar's average true range, a measure of volatility, over 14 days was used as a stop loss, and the target was set at 1.5 times the average true range. The position was closed at the 10th session close. If the price moved up by one average true range, half of the position was sold and the stop loss was moved to break even. If the price moved down, the same rules were applied. A gap stop, which is a stop loss that is triggered when the price gaps, or jumps, above or below a certain level, was used to determine the open and stop. All prices were adjusted for dividends and splits, which are changes to the stock's price or structure. A 5 basis point fee, which is 0.05%, and 5 basis points of slippage, which is the difference between the expected and actual price of a trade, were applied to each fill, which is the execution of a trade. The position size was continuously adjusted, with an initial cash amount of 100 and no leverage, which is the use of borrowed money to increase potential returns. The results were reconstructed using the vectorbt exact fill method, which is a way of simulating trades, and were found to be accurate within 3.27e-13 percentage points. The drawdown, which is the peak-to-trough decline in value, was calculated using synthetic per-day fill and close slots, rather than the actual 15-minute path. The positions were normalized to full allocation for comparison, but this is not a proposal. A bootstrap, which is a statistical method for estimating the variability of a result, was run 10,000 times with a seed of 90802, and the 95% mean interval was found to be between -0.384520 and 0.171548%. The historical profit and loss sizing was only used for training, and failures in training and holdout were kept at zero, even if they were positive in 2026. The data source was https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper was https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. This is not an exact leveraged betting against beta strategy. The QLV90802 metrics have 252 observations, and the study was deterministic, meaning that it was not random. The ledgers are local and can be found in the deterministic study.py file.
A frozen note, number 877, contains information about the QLV90802 USMV trading strategy. The strategy was trained using data from 2016 to 2020, and data from 2013 to 2015 is unavailable. The years 2021 to 2025 were used as a holdout period to test the new rules, but the market conditions during this time were also historically inspected. The year 2026 was used to complete the cohorts. The strategy is based on a monthly rule where the first session open after the previous month's close must be greater than the 100-day moving average. The previous bar's average true range over 14 days is used as a stop, with a target of 1.5 times the average true range, and the trade is exited at the 10th session close. A stop is also placed at the opening price if there is a gap, and the target is capped. The strategy does not allow partial profit and does not use leverage. The daily returns are adjusted for a 5 basis point fee and 5 basis point slippage for each fill. The strategy uses fractional continuous sizing, starting with 100 units of cash and one position. The vectorbt exact fill reconstruction agrees with independent compounding within 1.15e-13 percentage points. The drawdown is calculated using synthetic per day fill and close slots, not the actual 15-minute path. The positions are normalized for full allocation for comparison purposes only. A bootstrap analysis with 10,000 iterations and a seed of 90802 was performed, resulting in a 95% mean interval of -0.717164% to 0.723895%. The historical profit and loss sizing is only used for training and holdout, and failures during this time keep the actual value at 0, even if there are positive counterexamples in 2026. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the author's paper is https://pages.stern.nyu.edu/~afrazzin/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf. This strategy is not an exact leveraged betting against beta strategy. The full 252 metrics observations for QLV90802 are available, and the deterministic study is based on the ledgers local in the study.py file.