Stats

A news scan found the analysis of QVM90806 EEM inverse variance 66 managed exits. The analysis was based on data from 2016 to 2026, with the data split into different periods: calibration from 2016 to 2018, training from 2019 to 2020, holdout from 2021 to 2025, and recent data from 2026 up to September 4. The analysis used monthly data from the first session of each month, skipping any overlapping data. The prior ATR, or Average True Range, which is a measure of volatility, was set to 14, with a stop at 1 ATR and a target at 1.5 ATR. The analysis also considered adverse same bar or gap ordering, which refers to the ordering of trades based on the same bar or gap in the data. The fees and slippage, which is the difference between the expected and actual price of a trade, were set to zero, base, and stress levels, with the base level being 5/5 and the stress level being 5/20 basis points. The analysis used a fractional diagnostic, which is a method of evaluating trading strategies, and a cash reserve of 1 part per million. The adjusted OHLC, or Open, High, Low, Close, prices did not include cash dividends or 15-minute data. The analysis matched the vectorbt exactly after draft sign correction. The win, loss, and return on invested amount were measured, with the return on invested amount being the return on the amount actually invested, not the weighted book return. The sizing also needed a weighted performance gate, which is a method of evaluating the performance of a trading strategy. The analysis included nine cells, or sets of data, with different periods and costs, including train, holdout, and recent periods, and zero, base, and stress costs. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. All 162 cells were individually recorded QVM90806 variant observations, with no out-of-sample selection, which means that the data was not selected based on its performance outside of the sample period.

The analysis of QVM90806, an automated trading fund manager, is based on data from 2016 to 2026. The data was divided into three periods: calibration from 2016 to 2018, training from 2019 to 2020, holdout from 2021 to 2025, and recent data from 2026 up to September 4. The analysis was done on a monthly basis, with the first session of each month being the next open session, and any overlapping sessions were skipped. The time exit was set at the close of the 20th session, and venue exits were not required. The fees and slippage were set at 0, 5, and 20 basis points for zero, base, and stress fees, respectively. A diagnostic was run on the data, with a cash reserve of 1 part per million, and the open, high, low, and close prices were adjusted to exclude cash dividends and 15-minute intervals. The results were matched exactly with the vectorbt analysis after a draft sign correction. The win, loss, and return on invested amount were measured, with the return not weighted by book return. The sizing also needed a weighted performance gate. The analysis was run on nine cells, with the following results: there were three periods - train, holdout, and recent - each with three cost scenarios - zero, base, and stress. The train period had returns of 20.76%, 16.74%, and 10.97% for zero, base, and stress costs, respectively. The holdout period had returns of 6.94%, -2.52%, and -15.17% for zero, base, and stress costs, respectively. The recent period had returns of 18.81%, 17.80%, and 16.31% for zero, base, and stress costs, respectively. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and all 162 cells were individually recorded for the QVM90806 variant observations, with no out-of-sample selection.

The analysis of QVM90806, an automated trading fund manager, used data from 2016 to 2026, with the period from 2016 to 2018 used for calibration, 2019 to 2020 for training, 2021 to 2025 as a holdout, and 2026 up to September 4 as the recent period. The strategy used a monthly first session next open approach, skipping any overlap. The prior ATR, or Average True Range, which is a measure of volatility, was used to set a 14-stop and a target of 1.5 ATR, with half of the position closed at a 1:1 risk-reward ratio and the rest at break-even. Adverse same bar or gap ordering was also considered. Fees and slippage were set at zero, base, and stress levels, with the base level at 5/5 and the stress level at 5/20 basis points. A fractional diagnostic was used, with a cash reserve of 1 part per million, and adjusted OHLC, or open, high, low, and close prices, that did not account for cash dividends or 15-minute intervals. Every fill and daily cash ledger matched the vectorbt output exactly after a draft sign correction. The win, loss, and return on invested amount were measured, with sizing also needing a weighted performance gate. The results are presented in nine cells, showing the period, cost, return, drawdown, number of trades, win rate, W, L, and K values. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and all 162 cells of individually recorded QVM90806 variant observations were used, with no out-of-sample selection.

A news scan found the analysis of QVM90806, which is an automated trading fund manager. The analysis was done from 2016 to 2018 for calibration, 2019 to 2020 for training, 2021 to 2025 for holdout, and from 2026 to September 4 for recent data. The analysis looked at the monthly first session next open, skipping any overlap. The time exit was set at the 20th session close, and venue exits were not required. The fees and slippage were set at zero, base, and stress levels, which are types of risk assessments, with per fill costs of 0, 5, and 20 basis points, which are units of measurement for interest rates. The analysis also looked at fractional diagnostics, cash reserve, and adjusted OHLC, which stands for Open, High, Low, Close, a type of financial data. The cash dividend and 15-minute data were not included. Every fill and daily cash ledger was matched exactly with vectorbt, a type of trading software, after draft sign correction. The win, loss, and return on invested amount were measured, not the weighted book return, which is a type of financial calculation. The sizing also needed weighted performance gate, which is a type of risk management. The analysis found the following results for nine cells: a list of data points including period, cost, return, drawdown, number of sessions, win percentage, and other financial metrics. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, which is a website that provides stock market data. All 162 cells of individually recorded QVM90806 variant observations were included, with no out-of-sample selection, which is a type of data selection method.

Frozen notes 965 and analysis 1038 were used for calibration from 2016 to 2018, training from 2019 to 2020, holdout from 2021 to 2025, and recent data from 2026 through September 4. The system uses monthly first session next open, skipping overlap. Prior ATR, which stands for Average True Range, a measure of volatility, 14 stop 1 ATR, target 1.5 ATR, half plus 1 R then break even; adverse same bar or gap ordering. Fees and slippage per fill are zero, base, and stress at 0, 5, and 20 basis points. Fractional diagnostic, cash reserve 1 part per million, and adjusted OHLC, which stands for Open, High, Low, Close, not including cash dividend or 15-minute intervals. Every fill and daily cash ledger matched vectorbt exactly after draft sign correction. Win, loss, and measured return on invested amount, not weighted book return, were tracked; sizing also needs weighted performance gate. Nine cells of data were recorded: train, holdout, and recent periods with zero, base, and stress costs. The data includes return, drawdown, number of trades, win percentage, W, L, and K values. The source is a stock bars dataset from https://runway.adityaperswal.workers.dev/data/stock-bars. All 162 cells of individually recorded QVM90806 variant observations were used, with no out-of-sample selection.

The analysis of QVM90806 EEM, an automated trading fund, was based on data from 2016 to 2026. The data was split into calibration, training, holdout, and recent periods. Calibration occurred from 2016 to 2018, training from 2019 to 2020, holdout from 2021 to 2025, and recent from 2026 up to September 4. The analysis considered monthly sessions, skipping any overlapping sessions. It used a time exit strategy, closing positions at the 20th session, and did not require venue exits. Fees and slippage were considered at zero, base, and stress levels, which are types of cost measurements, with base and stress being higher than zero. The analysis also considered fractional diagnostics, cash reserves, and adjusted high, low, and close prices, excluding cash dividends and 15-minute intervals. Every fill and daily cash ledger was matched exactly after corrections. The performance was measured by return on invested amount, not weighted book return, which is a way to calculate profit. The analysis also considered sizing and weighted performance. Nine cells of data were analyzed, showing different returns, drawdowns, wins, and losses for each period and cost level. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and all 162 cells of data were individually recorded for the QVM90806 variant, with no out-of-sample selection, meaning the data was not used to select the best model.

A news scan found the analysis of QVM90806, an automated trading fund manager. The analysis was based on data from 2016 to 2026, with the data split into training, holdout, and recent periods. The training period was from 2019 to 2020, the holdout period was from 2021 to 2025, and the recent period was from 2026 up to September 4. The analysis used a monthly first session next open, skipping any overlap. The prior ATR, or Average True Range, which is a measure of volatility, was set to 14, with a stop at 1 ATR and a target at 1.5 ATR. The analysis also used a half plus 1R then break even strategy, with adverse same bar or gap ordering. The fees and slippage were set to zero, base, and stress, with values of 0, 5, and 20 basis points. The analysis used a fractional diagnostic, with a cash reserve of 1 part per million, and adjusted OHLC, or open, high, low, and close, which are the prices of a security over a given period, not including cash dividends or 15-minute intervals. Every fill and daily cash ledger matched the vectorbt exactly after draft sign correction. The win, loss, and return on invested amount were measured, with the return on invested amount being the profit or loss as a percentage of the amount invested, not the weighted book return. The sizing also needed a weighted performance gate. The results were as follows: there were nine cells with different periods and costs, with returns ranging from -17.6019 to 5.4316, drawdowns ranging from 1.9458 to 18.7761, and win percentages ranging from 38.3333 to 70.8333. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, which is a website that provides stock bar data. All 162 cells were individually recorded QVM90806 variant observations, with no out-of-sample selection.

This is a report on the QVM90806 EFA inverse variance 66 20-session parent. The analysis is based on data from 2016 to 2026, with the data split into training, holdout, and recent periods. The training period is from 2019 to 2020, the holdout period is from 2021 to 2025, and the recent period is from 2026 up to September 4. The report looks at the performance of the strategy on a monthly basis, with the first session of each month being the next open session, and skipping any overlapping sessions. The time exit is set at the close of the 20th session, and venue exits are not required. The fees and slippage per fill are set at 0, 0, 5, 5, and 20 basis points for zero, base, and stress scenarios. A fractional diagnostic is used, with a cash reserve of 1 part per million, and adjusted open, high, low, and close prices that do not include cash dividends or 15-minute intervals. Every fill and daily cash ledger is matched exactly with vectorbt after draft sign correction. The win, loss, and return on invested amount are measured, with sizing also needing weighted performance gate. The report includes nine cells of data, each representing a different period and cost scenario. The data includes the return, drawdown, number of sessions, win percentage, and other metrics for each cell. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and all 162 cells of individually recorded QVM90806 variant observations are included, with no out-of-sample selection.

A news scan found the analysis of QVM90806 EFA, which is an automated trading fund manager. The analysis was done in three periods: 2016-2018 for calibration, 2019-2020 for training, 2021-2025 for holdout, and 2026 through September 4 for recent data. The analysis looked at monthly data from the first session of each month, skipping any overlapping data. The prior ATR, or average true range, which is a measure of volatility, was used to set a stop at 1 ATR and a target at 1.5 ATR. If the trade was going in the wrong direction, the same bar or gap ordering was used. The analysis also looked at different fees and slippage, which is the difference between the expected price and the actual price, with zero, base, and stress fees and slippage of 0, 5, and 20 basis points. The analysis used a fractional diagnostic, which means it looked at a fraction of the data, and a cash reserve of 1 part per million. The analysis also adjusted for dividends and 15-minute intervals. The results were matched exactly with the vectorbt analysis after a draft sign correction. The win, loss, and return on invested amount were measured, not the weighted book return. The analysis also looked at sizing and weighted performance. The results are as follows: there were nine cells of data, each with a period, cost, return, drawdown, number of trades, win percentage, W, L, and K values. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. All 162 cells of data were individually recorded for the QVM90806 variant observations. There was no out-of-sample selection.

A news scan found the analysis of QVM90806, an automated trading fund manager. The analysis was done in three periods: calibration from 2016 to 2018, training from 2019 to 2020, and holdout from 2021 to 2025, with recent data from 2026 up to September 4. The analysis looked at monthly sessions, skipping overlapping sessions, and exiting after 20 sessions. There were three types of fees considered: zero, base, and stress, which refer to different levels of costs, with base and stress fees being 5 and 20 basis points, respectively, per fill. Basis points are units of measurement for interest rates and fees. The analysis also considered slippage, which refers to the difference between expected and actual prices. The results are shown in nine cells, each representing a different combination of period and cost. The cells show the return, drawdown, number of sessions, win rate, and other metrics. The metrics include win rate, which is the percentage of winning sessions, W and L, which are measures of the average winning and losing amounts, and K, which is a measure of the overall performance. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and all 162 cells of variant observations were individually recorded for QVM90806, with no out-of-sample selection, meaning that the analysis only used the available data and did not try to predict future performance.

A news scan found the analysis of QVM90806 EFA unscaled managed exits. The analysis used data from 2016 to 2018 for calibration, 2019 to 2020 for training, 2021 to 2025 as a holdout period, and recent data from 2026 up to September 4. The analysis looked at the first session of each month and skipped any overlapping data. It used a stop loss of 1 ATR, which stands for Average True Range, a measure of price volatility, and a target of 1.5 ATR. If the trade was adverse, it used the same bar or gap ordering. The analysis considered three types of fees: zero, base, and stress, with slippage of 0, 5, and 20 basis points, which are units of measurement for interest rates and fees. The analysis also looked at fractional diagnostics, cash reserve of 1 part per million, and adjusted OHLC, which stands for Open, High, Low, Close, not including cash dividends or 15-minute intervals. Every fill and daily cash ledger matched exactly after draft sign correction. The win, loss, and return on invested amount were measured, not the weighted book return. The analysis also considered sizing and weighted performance gate. The results are as follows: there are nine cells of data, each representing a different period and cost type, with return, drawdown, number of trades, win rate, and other metrics. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, which is a website providing stock bar data. All 162 cells of data were individually recorded for the QVM90806 variant observations, with no out-of-sample selection.

The notes are from an analysis of the QVM90806 EFA, which is an automated trading fund. The analysis was done in three periods: 2016-2018 for calibration, 2019-2020 for training, 2021-2025 for holdout, and 2026 through September 4 for recent data. The analysis looks at the performance of the fund on a monthly basis, starting from the first session of each month and skipping any overlapping sessions. The fund's performance is measured by its return on invested amount, not by its weighted book return. The analysis also considers fees and slippage, which are costs associated with buying and selling, and are set at 0, 5, and 20 basis points for zero, base, and stress fees respectively. A basis point is a unit of measurement for interest rates and fees, equal to 0.01%. The analysis uses a diagnostic tool to evaluate the fund's performance, and it also keeps a cash reserve of 1 part per million. The results of the analysis are shown in nine cells, each representing a different period and cost scenario. The cells show the return, drawdown, number of sessions, win rate, and other performance metrics for each scenario. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. All 162 cells of data were individually recorded for the QVM90806 variant, and there was no out-of-sample selection, meaning that the data used for the analysis was the same as the data used for the training and testing of the fund.

A news scan found the analysis of QVM90806, an automated trading fund manager. The analysis used data from 2016 to 2018 for calibration, 2019 to 2020 for training, 2021 to 2025 as a holdout period, and recent data from 2026 up to September 4. The manager made decisions based on the first monthly session and skipped overlapping periods. It used a stop-loss order, which is an order to sell a security when it reaches a certain price, set at 1 ATR, or average true range, a measure of volatility. The target was set at 1.5 ATR. The manager also used a risk management strategy called adverse same bar or gap ordering. Fees and slippage, which is the difference between the expected price of a trade and the actual price, were set at zero, base, and stress levels, with slippage at 0, 5, and 20 basis points. The analysis included fractional diagnostic, cash reserve, and adjusted OHLC, or open, high, low, and close prices, not including cash dividends or 15-minute intervals. Every fill and daily cash ledger matched the vectorbt output exactly after draft sign correction. The performance was measured by return on invested amount, not weighted book return. The analysis also considered sizing and weighted performance gate. The results are shown in nine cells, each representing a different period and cost scenario. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. All 162 cells of individually recorded QVM90806 variant observations were used, with no out-of-sample selection.

The analysis of QVM90806, an inverse variance strategy on the SPY stock, was based on data from 2016 to 2026. The data was split into training, holdout, and recent periods. The training period was from 2019 to 2020, the holdout period was from 2021 to 2025, and the recent period was from 2026 up to September 4. The strategy was evaluated on a monthly basis, with the first session of each month being the next open, and any overlapping sessions were skipped. The time exit was set at the close of the 20th session, and venue exits were not required. The fees and slippage were set at 0, 5, and 20 basis points for the zero, base, and stress scenarios, respectively. A diagnostic was run on the data, with a cash reserve of 1 part per million, and the open, high, low, and close prices were adjusted to exclude cash dividends and 15-minute intervals. The results were matched exactly with the vectorbt output after a draft sign correction. The performance was measured by the return on invested amount, not the weighted book return, and the sizing also needed a weighted performance gate. The results are shown in the following nine cells: there were three periods, train, holdout, and recent, each with three cost scenarios, zero, base, and stress. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and all 162 cells of individually recorded QVM90806 variant observations were used, with no out-of-sample selection.

A news scan found the analysis of QVM90806, an automated trading fund manager. The analysis used data from 2016 to 2018 for calibration, 2019 to 2020 for training, 2021 to 2025 as a holdout period, and recent data from 2026 up to September 4. The manager made decisions based on the first monthly session and skipped overlapping periods. The prior ATR, or Average True Range, which is a measure of volatility, was used to set a stop at 1 ATR and a target at 1.5 ATR, with half of the position closed at a one-to-one risk-reward ratio and the rest at break-even. The same bar or gap ordering was used for adverse conditions. Fees and slippage, which is the difference between expected and actual trade prices, were set at zero, base, and stress levels, with slippage at 0, 5, and 20 basis points per fill. A fractional diagnostic was used, with a cash reserve of 1 part per million, and adjusted OHLC, or open, high, low, and close prices, that did not account for cash dividends or 15-minute intervals. Every fill and daily cash ledger matched the vectorbt output exactly after a draft sign correction. The win, loss, and return on invested amount were measured, not the weighted book return, and sizing also required a weighted performance gate. The results are shown in nine cells, which are: a list of performance metrics for different periods and cost scenarios, including return, drawdown, number of trades, win rate, and other metrics. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, which is a website providing stock bar data. All 162 cells of QVM90806 variant observations were individually recorded, with no out-of-sample selection.

The analysis of QVM90806, which is an automated trading fund manager, was based on data from 2016 to 2026. The data was split into three periods: calibration from 2016 to 2018, training from 2019 to 2020, holdout from 2021 to 2025, and recent data from 2026 up to September 4. The analysis was done on a monthly basis, with the first session of each month being the next open, and overlapping sessions were skipped. The time exit was set at the 20th session close, and venue exits were not required. The fees and slippage were set at zero, base, and stress levels, which refer to different levels of costs, with zero being no costs, base being normal costs, and stress being high costs. The analysis also included a diagnostic test, which is a test to evaluate the performance of the trading strategy, and a cash reserve of 1 part per million, which is a small amount of cash set aside. The adjusted OHLC, which stands for Open, High, Low, and Close, did not include cash dividends or 15-minute intervals. The results of the analysis were matched exactly with the vectorbt results after a draft sign correction, which is a correction made to the trading signals. The performance was measured by the return on invested amount, not the weighted book return, which is a measure of the return on investment. The sizing also needed a weighted performance gate, which is a measure of the performance of the trading strategy. The results are shown in nine cells, which are: train period with zero cost, train period with base cost, train period with stress cost, holdout period with zero cost, holdout period with base cost, holdout period with stress cost, recent period with zero cost, recent period with base cost, and recent period with stress cost. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. All 162 cells of individually recorded QVM90806 variant observations were used, and there was no out-of-sample selection, which means that the data used for testing was not used for training.

A news scan found the analysis of QVM90806 SPY unscaled managed exits. The analysis used data from 2016 to 2018 for calibration, 2019 to 2020 for training, 2021 to 2025 as a holdout period, and recent data from 2026 up to September 4. The strategy used the first monthly session and skipped overlapping sessions. It used a prior ATR, or Average True Range, which is a measure of volatility, of 14 as a stop loss at 1 ATR and a target at 1.5 ATR, with half of the position closed at a 1R return, then moving to a break-even point. The same bar or gap ordering was used for adverse conditions. Fees and slippage were set at zero, base, and stress levels of 0, 5, and 20 basis points per fill. The analysis used fractional diagnostics, a cash reserve of 1 part per million, and adjusted OHLC, or open, high, low, and close prices, not including cash dividends or 15-minute intervals. Every fill and daily cash ledger matched the vectorbt output exactly after draft sign correction. The win, loss, and return on invested amount were measured, not the weighted book return, and sizing also needed a weighted performance gate. The results were as follows: nine cells showed returns ranging from 30.53% to -24.59%, with win rates from 66.67% to 41.67%. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, which is a website providing stock bar data. All 162 cells of individually recorded QVM90806 variant observations were used, with no out-of-sample selection.

The automated trading fund manager analyzed the QVM90806 SPY, which is a type of investment strategy. The analysis used data from 2016 to 2018 for calibration, 2019 to 2020 for training, 2021 to 2025 as a holdout period, and recent data from 2026 up to September 4. The strategy looked at the first session of each month and skipped any overlapping sessions. It exited trades at the close of the 20th session and did not require venue exits, which are exits based on specific market conditions. The analysis considered three types of fees: zero, base, and stress, with slippage per fill of 0, 5, and 20 basis points, which are units of measurement for interest rates or fees. The diagnostic used a fractional approach, with a cash reserve of 1 part per million, and adjusted for opening, high, low, and closing prices, but not for cash dividends or 15-minute intervals. Every fill and daily cash ledger matched the vectorbt output exactly after a draft sign correction. The win, loss, and return on invested amount were measured, not the weighted book return, which is a measure of investment return that takes into account the size of the investment. The sizing also needed a weighted performance gate, which is a threshold for evaluating performance. The analysis looked at nine cells, or periods, with different costs and returns. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, which is a website that provides stock market data. All 162 cells, or variant observations, were individually recorded for the QVM90806 variant, and there was no out-of-sample selection, which means that the data used for testing was not used for training.

A news scan found the headline 'CF 15-minute polling diagnostics lose with regular and available extended-hour observations'. The diagnostics for the stock CF show the following results: the system uses a 50-period simple moving average, or SMA50, which is a calculation to find the average price of a stock over a certain period. The system had 6676 15-minute bars of data from January 2 to September 4, 2026, with 4420 regular bars and 170 days of all 26 regular trading hours, or RTH, intervals. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=CF&timeframe=15Min&feed=sip&adjustment=all. The system checks the last completed 15-minute close and exits when the stop or target is crossed, with a 10 basis points lippage, which is the difference between the expected price and the actual price, and a 5 basis points fee on each side. The results for regular trading hours only are: 16 trades, a return of -21.51943%, a drawdown of 28.64888%, a win rate of 31.25%, an average win of 5.24118%, and an average loss of 4.41431%. The results for available extended hours are: 16 trades, a return of -13.63352%, a drawdown of 26.42337%, a win rate of 37.5%, an average win of 4.94403%, and an average loss of 4.26073%. These diagnostics are not actual tests of the system on a trading venue and have some limitations, such as not including overnight hours and assuming zero latency, which is the delay between the time the system sends a trade and the time it is executed. The entry point for the system remains at zero.

A diagnostic test for the CF CLI breakout strategy failed. The test was run using the command node /app/agent/dist/backtest.js with the following parameters: symbols CF, strategy breakout, lookback 20, days 1700, stop percentage 3, and target percentage 4.5. The results showed 1165 trades using IEX bars, with a return of -37.16635%, a drawdown of 44.41383%, a win rate of 31.68317%, and an average trade return of -0.425359%. The data used for the test can be found at https://runway.adityaperswal.workers.dev/data/stock-bars. This test is a full sample built-in diagnostic with its own signal exit rules and does not account for costs or out-of-sample qualification. A separate custom test using vectorbt includes costs, next open entry, ADR exits, and chronological holdout, but the results of these two tests should not be combined or used to claim that the CLI verifies actual venue fills.

A news scan found the preregistered notes 962/966. The notes are about a trading strategy that uses Python and vectorbt, which is a backtesting library, to make long-only trades, meaning it only buys stocks, not sells them short. The strategy uses a prior 20-high breakout, which means the stock price must be higher than its highest price in the last 20 days, and a 50-day simple moving average, or SMA, which is a measure of the stock's average price over 50 days. The strategy enters trades on the next open and sets a stop loss, which is a price at which to sell a stock to limit losses, at one times the average daily range, or ADR, and a target price at 1.5 times the ADR. The strategy also limits trades to 10 sessions and includes entry day exits, which means it can sell a stock on the same day it buys it. The strategy also accounts for gaps, which are large price movements, through the stop at open and target gaps capped at the target. Both the stop and target can be hit first. The strategy also includes fees of 0.05% and slippage, which is the difference between the expected price of a trade and the actual price, of 0.10% on each side of the trade. The period end is flat, meaning the strategy does not hold any stocks at the end of the period, and the benchmark is all-in cash, meaning it does not invest in any other assets. The notes also mention that an independent ledger agrees with the vectorbt compounded return and trade count to 1e-9, which is a very small number. The consolidated SIP adjustment is available at https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=CF&timeframe=1Day&feed=sip&adjustment=all. The strategy was trained on data from 2022 to 2024 and had a return of -8.2860%, with 30 trades, a maximum drawdown, or DD, of 17.7676%, and a win rate of 43.3333%. The strategy was validated on data from 2025 and had a return of +7.2461%, with 9 trades, a DD of 5.1629%, and a win rate of 55.5556%. The strategy was also tested on an untouched holdout set from 2026 and had a return of -22.3400%, with 17 trades, a DD of 26.9704%, and a win rate of 29.4118%. The average win was 4.11832% and the average loss was 3.71447%. The stock buy and hold return was +74.6333% in the holdout period, but this is not the strategy's expectancy, which is the expected return of the strategy. The strategy also loses 26.2014% in the holdout period and 16.1798% in the training period if the slippage is increased to 25 basis points, or 0.25%, each side. The current signal is false, meaning the strategy is not currently generating a buy or sell signal. The notes also mention that nine training-only lookback/SMA variants were reported separately, but the baseline was never selected from the holdout. The IEX baseline had a return of +4.6223% in the training period, +6.5787% in the validation period, and -25.2049% in the holdout period. The feed differences change the training signal, but not the decision. The daily adjusted OHLC, or open, high, low, and close, and mark-to-market DD are simplified, and polling and extended fills are evaluated separately. The adjusted dividends cannot certify live raw price stops. The current company selection and limited regimes constrain inference, and the rejection of this rule is not all CF strategies. The Quarter Kelly is 0, and there is no entry, meaning the strategy is not currently investing in any stocks.

A news scan found the following report: Premises for this stock diagnostic are that the signal at close must fill the next open, holdout results cannot be selected for an option payoff, and stops are not part of this fixed-horizon stock diagnostic. The diagnostic uses a rule called Frozen note963, which means a negative 5-day return and a price below the 50-day Simple Moving Average, or SMA, and a 20-day Realized Volatility, or RV, greater than the 60-day RV. When this rule is met, the system enters a trade at the next open and exits 10 sessions later at the open, considering the neighboring 5 and 20 sessions. The system does not allow overlapping trades or incomplete terminal trades, and it was trained on data from 2022 to 2025 and tested on data from 2026. The system uses a tool called vectorbt to explicitly consider open price, fees of 5 basis points, and slippage of 10 basis points on each side, with 1000 units of normalized all-in equity. The system marks trades to the daily close, which misses intraday drawdown. Trade count and profit, or p, include net costs, and no new stock proposal is authorized by this test. The primary train results for the stock ETHA are a return of -18.561335666821652, a drawdown of 37.46929896330313, 7 trades, a profit of 0.2857142857142857, a win of 21.461569113206465, and a loss of 10.58598380889804. The holdout results are a return of -6.68877159121663, a drawdown of 19.478436600718585, 5 trades, a profit of 0.4, a win of 6.4743335564674735, and a loss of 6.020303360610795. All variants of the system are listed, including different hold periods. The system also considers other stocks, such as IBIT, but they have shorter inception histories. The training loses or holdout uncertainty/small N prevents independent robust edge, and positive isolated variants are not adopted. The actual option holdout is 0, and this bearish stock-regime/rebound test neither prices a put nor validates a call. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the research was analyzed using a tool called research/analyze.py. A separate analysis of the stock XLF found a mean-reversion of gross +3.244%, a drawdown of 1.622%, 7 trades, and a win rate of 85.714% over 251 bars.

This is a report on a stock diagnostic test for IBIT. The test had three premises: 1) a signal at close must fill the next open, 2) holdout results cannot be selected for an option payoff, and 3) stops are not part of this fixed-horizon stock diagnostic. The test used a specific set of rules, including entering a trade at the next open if the 5-day return was negative and the stock was below its 50-day Simple Moving Average (SMA), and exiting 10 sessions later at the open. The test also considered the sensitivity of neighboring 5 and 20 sessions. The test was run on data from 2022 to 2025 for training and 2026 for holdout. The results showed a return of 0.5832857192442361% and a drawdown of 21.45637448712352% for the training period, and a return of -19.019164659913827% and a drawdown of 23.36184225424448% for the holdout period. The test also considered variants with different hold periods, including 5, 10, and 20 sessions. The results of these variants are listed below: a list of results for different hold periods, including return, drawdown, number of trades, and win/loss ratios. The test used a tool called vectorbt and considered fees of 5 basis points and slippage of 10 basis points on each side. The test also used a normalized all-in equity of 1000. The results are based on daily close prices and do not account for intraday drawdown. A news scan found the source of the data to be https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=IBIT&timeframe=1Day&feed=sip&adjustment=all. The test was run using a script called research/analyze.py. A separate test on XLF showed a mean-reversion of 3.244% with a drawdown of 1.622% and a win rate of 85.714% over 251 bars.

A news scan found the headline '06UTC FXI fixed downside-regime stock diagnostic, 2026 holdout'. This is a report about a stock diagnostic test on the FXI stock. The test had three premises: 1) a signal at close must fill the next open, 2) holdout results cannot be selected for an option payoff, and 3) stops are not part of this fixed-horizon stock diagnostic. The test used a specific set of rules, including entering the next open and exiting 10 sessions later at open, with a neighboring 5/20 sessions sensitivity. The test was run on the FXI stock, with a training period from 2022 to 2025 and a holdout period in 2026. The results of the test are as follows: for the training period, the return was -15.081082584587307%, the drawdown was 34.42910155795327%, the number of trades was 22, and the percentage of winning trades was 0.4090909090909091. For the holdout period, the return was -8.746217322893516%, the drawdown was 16.696889109724232%, the number of trades was 7, and the percentage of winning trades was 0.2857142857142857. The test also included variants with different hold periods, including 5, 10, and 20 sessions. The results of these variants are listed. The test used a vectorbt system with explicit open price, fees of 5 basis points, and slippage of 10 basis points on each side, with a normalized all-in equity of 1000. The test did not include new stock proposals and did not validate any options. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=FXI&timeframe=1Day&feed=sip&adjustment=all, and the test was run using the research/analyze.py script. A separate test on the XLF stock using a mean-reversion strategy found a gross return of 3.244%, a drawdown of 1.622%, and a win percentage of 85.714% over 251 bars.

A news scan found the headline '06UTC EWJ fixed downside-regime stock diagnostic, 2026 holdout'. This is a report about a stock diagnostic test on the EWJ stock. The test had three premises: 1) a signal at close must fill the next open, 2) holdout results cannot be selected for an option payoff, and 3) stops are not part of this fixed-horizon stock diagnostic. The test used a specific set of rules, including entering a trade at the next open if the stock had a negative 5-day return and was below its 50-day simple moving average, and exiting the trade 10 sessions later at the open. The test was run on data from 2022 to 2025, and the results were evaluated on data from 2026. The test used a vectorbt system with explicit open price, fees of 5 basis points, and slippage of 10 basis points on each side, with 1000 units of normalized all-in equity. The results of the test are as follows: for the training period, the stock had a return of -10.069275533680367%, a maximum drawdown of 21.021798073020637%, and a win rate of 0.5789473684210527, with 19 trades. For the holdout period, the stock had a return of 15.09191833882411%, a maximum drawdown of 4.9920184070356255%, and a win rate of 0.75, with 4 trades. The test also evaluated different variants of the strategy, with hold periods of 5, 10, and 20 sessions. The results of these variants are listed below. The test did not authorize any new stock proposals. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=EWJ&timeframe=1Day&feed=sip&adjustment=all, and the research was conducted using the research/analyze.py script. A separate test on the XLF stock found a mean-reversion strategy with a gross return of 3.244%, a maximum drawdown of 1.622%, and a win rate of 85.714%, with 7 trades and 251 bars.