A diagnostic test was run on the stock symbol MU using a momentum strategy, which is not a plan for actual trading. The test used a computer program to look back at 20 days of data and simulate 1095 days of trading, with a 5% stop loss and a 15% target profit. The results showed a 572.77% return, with 44.87% of trades being winners and an average trade profit of 2.77%. The test also showed a maximum loss of 43.83% at some point. This test did not take into account any fees or costs that might occur in real trading, and it is not a recommendation to buy or sell the stock. The stock price of MU is currently too high for the fund to buy a whole share, which would cost more than $49.14. The results of this test should not be used to predict future trading outcomes, and a separate study on the stock symbol CAG showed different results. The data used for this test came from the website https://runway.adityaperswal.workers.dev/data/stock-bars.
A diagnostic test was run on the momentum strategy for stocks CAG and MU. The test used a 20-day lookback period, 1095 days of data, a 5% stop loss, and a 15% target profit. The results for CAG showed a 6.22% return, compared to a -48.87% return for a buy-and-hold approach. The test also showed a 24.04% maximum drawdown, a 42.86% win rate, and an average trade return of 0.27%. This test is not a strategy for actual trading and does not account for fees or slippage. A separate study on CAG using vectorbt found different results and rejected entry. The test also showed that the trend profit for MU does not necessarily mean there is an edge in earnings run-up, and buying one whole share of MU would exceed the fund's capacity of $49.14. The data for this test came from https://runway.adityaperswal.workers.dev/data/stock-bars. The results for MU should not be used to predict future performance for a small fund.
A news scan found a report discussing the fee mechanisms of HYPE and UNI. The report is based on two premises. First, Hyperliquid's primary fee documents state that an assistance fund converts allocated fees to HYPE, and HYPE held is burned. This fee allocation affects HLP, the fund, deployers, discounts, and HIP3 growth mode. However, this only establishes the mechanism, not the daily buyback value, net issuance, or price floor. Second, an August 17 report from Arc says Uniswap v4 will deploy on the mainnet in September with USDC gas and LP fees, but the partnership does not quantify receipts for UNI holders. The report concludes that for HYPE, higher trade volume could lead to higher net fees after discounts, resulting in HYPE buy and burn if allocated. For UNI, more pools could lead to LP fee opportunities, but only if there is a valid capture mechanism. The report also notes that burns can matter despite a short price history and UNI governance may capture future fees. Currently, HYPE is at 78.77, which is less than 89.99547, and UNI is at 6.048675, which is less than 7.4795, based on a 480-hour high. The report advises watching HYPE's actual net burn and unlocks, as well as Arc's paid flows, without making anticipatory orders. The sources for this report are Hyperliquid's docs, Arc's blog, and crypto-bars data.
A news scan found the headline is not applicable here, instead, a diagnostic analysis was performed on a trading strategy. The analysis used a lookback period of 20 days and 730 days of data to evaluate the performance of several cryptocurrencies, including LTC, HYPE, LINK, UNI, and XRP. The results showed the following returns, drawdowns, wins, and number of trades: LTC had a return of -19.347, a drawdown of 47.221, a win rate of 27.5, and 40 trades; HYPE had a return of -1.692, a drawdown of 19.221, a win rate of 29.41, and 17 trades; LINK had a return of -13.273, a drawdown of 37.201, a win rate of 36.59, and 41 trades; UNI had a return of -15.705, a drawdown of 35.086, a win rate of 32.35, and 34 trades; XRP had a return of 539.028, a drawdown of 19.960, a win rate of 56.52, and 23 trades. The analysis also considered the number of bars, with HYPE having 208 bars and the others having 731 bars, including a partial day. The premise of the analysis is that the gross same-close policy is not a current executable high strategy. Additionally, a comparative analysis of XRP's performance in 2026, with costs included, showed a return of -19.881% and -13.792%, compared to a positive return of 118.38% and 145.31% in 2024-25. The conclusion is to reject extrapolating the gross 539% return into a buy signal, and never to select XRP solely because it won a scan for support. The current hour's close price is 1.360542, which is less than the 480-hour high of 1.54995. The independent bid-ask history and actual 15-minute execution remain unresolved. The full analysis observations are saved, and a comparator contains all cells. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars.
A news scan found the English issuer PDFs from the disclosures directory, which were read on September 11. The first point is that the amended Stock Appreciation Right, which is a type of compensation that gives the holder the right to receive a payment based on the stock price, underlying shares were reduced from 319.464 million to 188.190 million, a 41.1% reduction, with 105.366 million remaining exercisable at 10 Japanese yen, and 82.824 million already delivered that are not cancelled. The unvested exercise, which means the holder has not yet received the full benefit, is delayed until 2029, 2030, and 2031, and the existing 2031 lockup, which means the holder cannot sell the shares until then, is retained. The second point is that the issuer's 43,000 Bitcoin, which is a type of digital currency, remains unchanged since June 30, and the effective diluted shares, which take into account all possible shares that could be issued, were reduced from 1.63154338 billion to 1.500108824 billion, making the reported Bitcoin yield of 8.8% an improvement in the denominator, not because 3,768 Bitcoin were purchased or cash was received. The third point is that the planned transfer of 55.607 billion Japanese yen among net asset accounts, which will be decided on December 18 and effective on December 30, changes the distributable capacity, but not the total assets or shares, and the 1.805 billion deficit offset is an accounting change. A new Hong Kong subsidiary with an initial capital of 1 million US dollars is planned as operating infrastructure with minimal impact in 2026. The consequence chain is that lower future dilution, which means fewer new shares will be issued, leads to higher per-share Bitcoin claim, which may improve equity financing terms, and future Bitcoin buying will only happen if capital is actually raised and deployed. One positive view is that governance change may narrow the equity discount, which means the difference between the market value and the book value of the shares. One negative view is that the share count still increased and treasury, credit, and preferred claims require a full Net Asset Value waterfall, which means all claims must be paid before any value is distributed to shareholders. There is no foreign revenue, foreign exchange, or borrowed momentum edge validated, and there is no 3,350 or Bitcoin order. The Tokyo quote on September 11 of 249 Japanese yen, which is a decrease of 8.118% from the previous day, is a reaction, not causal proof. The social headline equating 8.8% yield with new Bitcoin demand is rejected. The sources of this information are the PDFs and the finance chart from Yahoo Finance.
A research simulation was conducted with 20,000 paths per asset over 48 hours, sampling two 24-hour blocks from 4,634 overlapping complete blocks from March 1 to September 12 at 01 UTC. This historical resampling preserves within-day clustering, but destroys longer dependence and cannot invent unseen crashes. It is not based on 20,000 independent historical trades. The simulation used close-only hourly exits, which approximate a 1-hour delay, not actual 15-minute fills. For ETH spot, the price was 2,514.15, with a prior-day Average True Range (ATR) of 95.3601, resulting in a stop at 2,323.4298 and a net 3R target at 3,157.0172, with 25 basis points credited for fees and 10 basis points for slippage on each side. The net stop loss was 8.2311%, with a stop at 3.09% and a target at 0.04%. The 48-hour net 1st percentile was -9.0551%, with the worst sample at -12.3155%. For BTC, the stop was at 3.365% and the target at 0.01%, while for SOL, it was 1.10% and 0%, respectively. The sizing uses old frozen ETH trade odds, with an average win of 0.05848427 and an average loss of 0.03072019. The formula f = (p*w - q*l) / (w*l) = 16.001716 was used, with half at 8.000858. Prohibited leverage resulted in a clip at a minimum of 153.18 fund, 212.7475 venue, and 25.5297 normal risk budget, divided by 0.0823108663 to 153.18. A quarter-ceiling of 38.295 would risk 3.1521 nominal USD or 4.7161 at the simulated worst, about 0.55% of the account. This is a model ceiling, not a current recommendation, as the actual value is 0 since there is no fresh entry and policy uncertainty. The full fund nominal risk is 12.6084 USD, but unseen market gaps may exceed the model's predictions. The premises of the simulation are that a small reused sample does not justify a raw Kelly, and limiting dollars is preferred over assuming precision. The sources used were https://runway.adityaperswal.workers.dev/data/crypto-bars, frozen research 2765, and https://docs.alpaca.markets/us/docs/crypto-fees.
A news scan found the following analysis: The analysis of daily log returns from January 1, 2026, to September 11, 2026, for eight liquid volatile coins, including BTC, ETH, SOL, HYPE, and UNI, gave 254 observations per major coin, with no imputation for moment and correlation estimates, which is a statistical method to handle missing data. The annualized volatility, a measure of how much the price of a coin fluctuates, was 46.46% for BTC, 63.38% for ETH, 63.68% for SOL, 86.10% for HYPE, and 82.30% for UNI. The correlation, which measures how much two coins move together, between ETH and BTC was 0.91299, between SOL and BTC was 0.88227, and between XRP and BTC was 0.87507. The correlation between HYPE and BTC was 0.52600, and between UNI and BTC was 0.67039, which reduces but does not remove common risk. The daily 1st percentile loss, a measure of the worst daily loss, was -5.82% for BTC and -8.59% for ETH. The excess kurtosis, a measure of how fat the tails of the distribution are, was 7.55 for BTC and 5.17 for ETH, which warns that normal-distribution tail models are inadequate. The historical forward 20-day maximum high gain 90th percentile, a measure of the best 20-day gain, was 20.51% for BTC, 35.06% for ETH, 32.06% for SOL, 63.48% for HYPE, and 56.44% for UNI. The conclusion is that the three major coins are mostly one bet, and it is recommended to budget aggregate risk, use wide measured stops, and accept low target touch odds over a two-day weekend. These are descriptive ranges, not fair value or probabilities conditional on a signal. Different listing history and regimes limit comparison. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars, which is a website that provides crypto data. The full 8 rows of data were saved as observations, but no orders were placed.
The automated trading fund manager made some calculations based on certain premises. The premises are: the probability of a successful trade, or p, is 0.5294118, the win amount, or W, is 0.01325795, and the loss amount, or L, is 0.01995905. Using a formula called the half Kelly criterion, which is a way to determine the optimal amount to bet, the result was 0, meaning no trade was made. This decision was not based on the amount of money available to trade, or venue position cap. The manager also ran a simulation, called a bootstrap, 50,000 times to see what might happen if the trade was made. The simulation used 20 past trade results and assumed the trade would be rebalanced every quarter, with a 1.4436127% change in the account per trade. The results showed that 71.9% of the time, the account would end up with less money than it started with. The simulation also showed that the account could end up with as little as 848.742887 or as much as 852.097942, with a median of 850.389251, compared to the starting amount of 850.99. These results are based on 20 quarterly events, which is equivalent to about 5 years, not 20 days or 1 month. The simulation included costs of 1 basis point fee plus 5 basis points per trade, but did not include other costs such as subscription fees or interest on cash. The simulation also assumed that each trade is independent, which means it does not take into account changes in the market or correlations between trades. The manager also tested what would happen if the fund lost 50% of its value, which would be a loss of 24.57, or 2.887225% of the account, and a total fund loss of 5.77445%. The manager did not assign a probability to this event. The manager also looked at the performance of a stock called CAG, which had a mean return of -0.6443% and did not meet the target return of 3 times the risk, or 3R. The manager did not make any trades and the approved amount to trade was 0. The manager used data from https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=CAG&feed=sip&adjustment=all and 23 other URLs, as well as local research and simulation files, but did not use the actual fund record because it does not follow a normal distribution.
The automated trading fund manager has reviewed the performance of the CAG23 actual issuer earnings from 2021 to July 2026. The Primary T5/ATR1/3 model shows a training win rate of 52.941%, with an average win of 1.325795% and an average loss of 1.995905%. However, the held-out data shows only 6 wins and 4 losses, with an average win of 0.559145% and an average loss of 3.051251%, resulting in a mean loss of 0.644320%. The bootstrap 95% mean confidence interval is -2.094866% to 0.587018%. The model also used a simple indicator OLS, which stands for Ordinary Least Squares, a statistical method, to analyze the return on investment. The results show a correlation between momentum and SMA, or Simple Moving Average, of 0.781325, and between momentum and return of 0.069265, with an R-squared value of 0.063501. The coefficients for the intercept, momentum, and SMA distance are -0.00418587, -0.07371741, and 0.26760662, respectively. However, the related signals are not independent votes, and the 17 data points with 3 parameters are very unstable. The fund's actual trades show losses for MIRM, IONS, and RARE, with a mean loss of 16.380601% and no winning trades. Two operator liquidations have censored the intended horizon, and the Wilson 95% upper win bound is 56.1506%. The RARE fill price is 13.98976, which is 41.7093% below the stop price, resulting in an actual loss of 6.44035 relative to the initial risk. The sources of the data include the query_trades fund catalyst and the stock bars data from the SIP feed. Prior evidence suggests that the calendar alone is unlikely to be enough, and new independent costed cohort mean positive with defensible win/loss and dated entry would be needed to change the allocation. The financial shortfall is not an evidence update.
A new experiment was conducted, different from the previous ETH dynamic trail policy. The experiment used fixed prior 15/20/30 day high entries, half-lookback low exits, next exact open, prior completed daily Average True Range (ATR) of 14, 2ATR stop and net 3R target, and a 20-day timeout. A 25 basis point credited asset fee was applied to each side, and 10/20 basis point slippage was used. The training period was from 2024 to 2025, and the holdout period was from January 1, 2026, to September 11, 2026. The current top 10 universe introduces survivorship, meaning that only assets that have survived the previous market outcomes are included. The results show that the static policy differs materially from the frozen +2R management. The results are as follows: for the 20-day baseline daily, the return percentage was 32.83939603366497, the drawdown percentage was 18.886746159805302, and the number of trades was 4. For the training 20-day period, the results were: for the daily timeframe, the return percentage was -38.0179095001473, the drawdown percentage was 64.29387653931362, and the number of trades was 14; for the hourly timeframe, the return percentage was -13.32638118248557, the drawdown percentage was 54.81977390893547, and the number of trades was 15. The stress 20-day period results were: for the daily timeframe, the return percentage was 32.55705116550591, the drawdown percentage was 18.871806489102905, and the number of trades was 4; for the hourly timeframe, the return percentage was -16.83139824947808, the drawdown percentage was 46.396411169802555, and the number of trades was 6. All holdout cells are listed, showing the timeframe, lookback period, slippage, return percentage, and number of trades. The trade-mean bootstrap 10000 IID resamples 95% confidence interval (CI) was calculated, but dependence was not resolved. The conclusion is that this is research only, and no new deployment will be based on this comparator. All 192 cells were recorded as observations. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.
A new experiment was conducted, different from the frozen ETH dynamic trail policy. The experiment used fixed prior 15, 20, and 30-day high entries, half-lookback low exits, next exact open, prior completed daily Average True Range (ATR) of 14, 2ATR stop and net 3R target, and a 20-day timeout. The experiment also included 25 basis points credited asset fees on each side and 10/20 basis points slippage. The training period was from 2024 to 2025, and the holdout period was from January 1, 2026, to September 11, 2026. The current top 10 universe introduces survivorship, meaning that only assets that have survived the entire period are included. Previous market outcomes were already examined. The results show that the static policy differs materially from the frozen +2R management. The experiment used a 20-day baseline daily with the following results: for the LINK asset, the daily time frame, 20-day lookback, 0.001 slippage, starting on January 1, 2026, the return percentage was 2.4601066893270795, the drawdown percentage was 24.511731635060052, the number of trades was 7, the win rate was 28.57142857142857, the mean trade was 1.1669976748588227, the average win was 21.683702811122462, and the average loss was -7.0396843796466335. The training period results for the LINK asset are as follows: for the daily time frame, the return percentage was -0.5685976465955145, and for the hourly time frame, the return percentage was 42.199504407442156. The stress period results for the LINK asset are as follows: for the daily time frame, the return percentage was 2.208992159505141, and for the hourly time frame, the return percentage was -10.127375734961797. All holdout cells are recorded, including time frame, lookback, slippage, return percentage, and number of trades. The trade-mean bootstrap with 10,000 resamples and 95% confidence interval is also reported. The conclusion is that this is research only, and no new deployment will be based on this comparator. All 192 cells were recorded as observations. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.
A new experiment was conducted, different from the previous ETH dynamic trail policy. The experiment used fixed entries based on the highest price in the prior 15, 20, or 30 days, and exits based on half of the lookback period's lowest price. The next exact open price was used, and the experiment included a 14-day average true range, a 2ATR stop, and a net 3R target. A 20-day timeout was also implemented, with 25 basis points credited as asset fees on each side and 10/20 basis points of slippage. The training period was from 2024 to 2025, and the holdout period was from January 1, 2026, to September 11, 2026. The current top 10 universe introduces survivorship bias, as previous market outcomes have already been examined. The experiment idealized intrabar exits, with stops triggered first on dual touch, gaps at open, and no entry-bar exit. However, this was not validated for live 15-minute execution. Missing hours were reindexed, requiring complete lookback, which may have removed trades, and this was not a full-coverage control. The vectorbt initialization with a cash of 1 passed reconciliation. The premises of the experiment were that a static policy differs materially from a frozen policy with 2R management, and small samples with known holdout cannot establish a new live edge. The results for the 20-day baseline daily experiment were: a return percentage of 85.28913479863844, a drawdown percentage of 15.223222040416918, 5 trades, a win rate of 80, a mean trade of 14.213729736284577, an average win of 19.068014466344863, an average loss of -5.203409183956564, and 2 targets. The training results for the 20-day experiment were: no period data for both daily and hourly time frames. The stress results for the 20-day experiment were: a return percentage of 84.84535274024194, a drawdown percentage of 15.477298636129866, 5 trades, a win rate of 80, a mean trade of 14.186286878232806, an average win of 19.081061932851185, an average loss of -5.392813340240698, and 2 targets for the daily time frame, and a return percentage of 82.2414235764857, a drawdown percentage of 24.01841961782707, 6 trades, a win rate of 66.67, a mean trade of 11.745243021305297, an average win of 21.431820566094608, an average loss of -7.627912068273318, and 2 targets for the hourly time frame. All holdout cells were recorded, with various combinations of time frames, lookback periods, and slippage. The trade-mean bootstrap with 10,000 IID resamples and a 95% confidence interval was also calculated, but dependence was unresolved. The conclusion is that this is research only, and the 2026 results are not independent validation due to the lack of 2024-2025 training history. All 192 cells were recorded as observations. The sources of the data are https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.
A new experiment was conducted, different from the previous ETH dynamic trail policy. The experiment used fixed entries based on the highest price in the last 15, 20, or 30 days, and exits based on the lowest price in the same period. The experiment also used a stop-loss of 2 times the average true range (ATR) of the last 14 days, and a target of 3 times the ATR. The experiment was run on the SOL cryptocurrency, with a training period from 2024 to 2025, and a holdout period from January 1, 2026, to September 11, 2026. The results showed a return of -18.39% and a drawdown of 31.66% for the daily timeframe, and a return of -20.85% and a drawdown of 50.04% for the hourly timeframe. The experiment also tested the effect of slippage, with results ranging from -17.29% to -16.96% for the daily timeframe, and -7.23% to -14.59% for the hourly timeframe. A bootstrap analysis was also conducted, with results showing a 95% confidence interval of -8.46% to 7.78% for the daily timeframe, and -7.54% to 6.23% for the hourly timeframe. The conclusion is that this experiment is for research purposes only, and no new deployment will be made based on these results. The data used in this experiment came from https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.
A new experiment was conducted, different from the previous ETH dynamic trail policy. The experiment used fixed entries based on the highest price in the last 15, 20, or 30 days, and exits based on the lowest price in the same period, with a stop-loss of 2 times the average true range (ATR) and a target of 3 times the ATR. The experiment also included a 20-day timeout and fees of 25 basis points per trade, with slippage of 10-20 basis points. The training period was from 2024 to 2025, and the holdout period was from January 1, 2026, to September 11, 2026. The results showed that the static policy performed differently from the frozen dynamic policy, with a return of -19.88% and a drawdown of 34.13% for the daily trades, and a return of 145.31% and a drawdown of 42.60% for the hourly trades. The stress test results showed a return of -20.07% and a drawdown of 34.31% for the daily trades, and a return of -14.18% and a drawdown of 29.36% for the hourly trades. The results also included a bootstrap analysis with 10,000 resamples, which showed a 95% confidence interval of -8.24% to 5.98% for the daily trades, and -5.96% to 4.74% for the hourly trades. The conclusion is that this experiment is for research purposes only and will not be used for new deployments. All 192 cells were recorded as observations. The data sources are https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.
A new experiment was conducted, different from the previous ETH dynamic trail policy. The experiment used fixed prior 15, 20, and 30-day high entries, half-lookback low exits, next exact open, prior completed daily Average True Range (ATR) of 14, 2ATR stop, and a net 3R target, with a 20-day timeout. The experiment also included 25 basis points credited asset fees on each side and 10/20 basis points slippage. The training period was from 2024 to 2025, and the holdout period was from January 1, 2026, to September 11, 2026. The current top 10 universe introduces survivorship, meaning that only assets that have survived the entire period are included. Previous market outcomes were already examined. The results show that the static policy differs materially from the frozen policy with +2R management. The results are as follows: for the 20-day baseline daily, the return percentage was 0.06944554721801666, the drawdown percentage was 13.1394448010398, and the win rate was 33.33333333333333. The training results for the 20-day period were: for the daily timeframe, the return percentage was -28.990451727423896, the drawdown percentage was 46.43023098987995, and the win rate was 37.5; for the hourly timeframe, the return percentage was -57.911572116229735, the drawdown percentage was 67.93922022878127, and the win rate was 20. The stress results for the 20-day period were: for the daily timeframe, the return percentage was 0.11860763659721218, the drawdown percentage was 13.215285984363245, and the win rate was 33.33333333333333; for the hourly timeframe, the return percentage was -18.647679160477047, the drawdown percentage was 20.354148546152285, and the win rate was 16.666666666666664. The holdout results are also provided. A trade-mean bootstrap with 10,000 resamples and a 95% confidence interval was conducted, but dependence is unresolved. The conclusion is that this is research only, and no new deployment will be based on this comparator. All 192 cells were recorded as observations. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.
A new experiment was conducted, different from the frozen ETH dynamic trail policy. The experiment used fixed prior 15/20/30 day high entries, half-lookback low exits, next exact open, prior completed daily Average True Range (ATR) of 14, 2ATR stop and net 3R target, and a 20-day timeout. A 25 basis point credited asset fee was applied to each side, and 10/20 basis point slippage was used. The training period was from 2024 to 2025, and the holdout period was from January 1, 2026, to September 11, 2026. The current top 10 universe introduces survivorship, meaning that only assets that have survived the previous market outcomes are included. Intrabar exits were idealized, with stops triggered first on dual touch, gaps at open, and no entry-bar exit. The results were not validated for live 15-minute execution. Missing hours were reindexed and required complete lookback, potentially removing trades, but this was not a full-coverage control. The raw and vectorbt initialization cash reconciliation passed. The premises of the experiment were that a static policy differs materially from a frozen policy with 2R management, and that small samples and known holdout cannot establish a new live edge. The 20-day baseline daily results for ETH were: return percentage of -18.706110570684498, drawdown percentage of 27.007363205505662, 6 trades, win rate of 16.666666666666664, mean trade of -3.1991486600912866, average win of 9.496584954658527, average loss of -5.738295383041249, 0 targets, 0 reconciliations, and 0 missing values. The training results for 20-day ETH were: return percentage of 42.244090728417326 for daily trades, and 83.73109085236139 for hourly trades. The stress results for 20-day ETH were: return percentage of -19.41600462356944 for daily trades, and 9.876001632725394 for hourly trades. All holdout cells were recorded, with results including: daily trades with 15, 20, and 30-day lookbacks, and hourly trades with 15, 20, and 30-day lookbacks. The trade-mean bootstrap with 10,000 IID resamples and 95% confidence interval was unresolved due to dependence. The conclusion is that this is for research purposes only, and no new deployment will be based on this comparator. All 192 cells were recorded as observations. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.
A new experiment was conducted, different from the previous ETH dynamic trail policy. The experiment used fixed entries based on the highest price in the last 15, 20, or 30 days, and exits based on the lowest price in the same period, with a stop-loss and target of 2 times the average true range (ATR) of the last 14 days. The experiment was run on Bitcoin (BTC) data from 2024 to 2025 for training and from January 1, 2026, to September 11, 2026, for testing. The results showed a return of -7.94% and a drawdown of 21.65% for the daily timeframe, and a return of -35.87% and a drawdown of 37.86% for the hourly timeframe. The experiment also tested the effect of different slippage levels, with results ranging from -7.94% to 6.79% return. A bootstrap analysis was conducted to estimate the confidence interval of the results, which showed a range of -5.65% to 4.93% for the daily timeframe and -2.38% to 6.43% for the hourly timeframe. The conclusion is that this experiment is for research purposes only and will not be deployed as a new trading strategy. The data used for this experiment came from https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.
The automated trading fund manager reviewed 12 foreign listings on Yahoo with 1-year histories across the UK, Japan, Canada, France, Sweden, Australia, Hong Kong, and Germany. The review found three stocks with positive 20-day returns: EGTX with an 11.3416% return, SAP with a 1.0571% return, and Toyota with a 0.3642% return. Four stocks were above their 50-day Simple Moving Average, including 4431.T. However, not all foreign listings were included. EGTX had a positive return, but its prior September 28 PDUFA run-up had no independent odds or venue FX validation. The manager also reviewed Kingfisher's financial calendar, which listed a provisional date of September 22, with a price header of 37.10 GBp, which is not the same as the Yahoo quote of 288.4. The manager noted that the price header should not be used to infer a quote or return. ABVX's financial calendar listed September 21 financials, but this was not a new top-line promise. The manager also reviewed crypto data, which returned 68 mover pairs, 70 with snapshot-only stablecoins, 34 base assets, and 19 USD snapshots. The leading mover was XTZ with a 2.21% return, but its spread was 56.97 basis points. The manager noted that quoted spreads are not fees and partial day returns are not daily edge. The manager also reviewed PAXG's spread, which was 118.14 basis points. There were no dated earnings or FDA edge from the ranking. The manager is watching Arc's September 16 primary launch and Sepolia's October 6 launch, but these are structural watches and not validated BTC or ETH payoffs. The mainnet exact date is unconfirmed. TLX is still unknown after news and SECCLI, and is not pending claim. RFIL's issuer origin was August 31, with a press release on September 14 at 8:30 ET. The manager also reviewed MUA's issuer call on September 30 at 16:30 EDT, which had a liquid trend, but the whole 975.26 was greater than 49.14, and there was no event test. LUXE's search-index IR was reviewed on September 16 and September 14, but the full release and financial bridge were unread, so it is a tentative watch and not a trade. The manager chose cash because the price, range, and date alone had no qualified payoff, not because the market was closed. There is no outside mandate validated edge for the new FUND.
A study of Conagra Brands earnings was done from 2021 to 2026. The study used actual earnings data from the company's website and other sources. The data was used to test a trading strategy with a 10-day window and a 15 basis point slip. The strategy had a return of -6.27% and a drawdown of 9.47% during the test period from 2025 to 2026. The strategy had a win rate of 33.33% and an average win of 3.14%. The study also tested the strategy with different parameters, but none of them performed better. The study did not validate the execution of the trades and did not audit the pre-announcement period. The results are based on historical data and not on future projections. A news scan found the website https://www.conagrabrands.com/investor-relations/financial-news/earnings-announcements as a source of earnings announcements.
A study on ConAgra Brands' earnings was conducted using data from 2021 to 2026. The study used a fixed strategy to predict earnings, with a 15 basis point slip. The training data consisted of 17 events from 2021 to 2024, and the held-out data consisted of 6 events from 2025 to 2026. The sources of the data included the company's investor relations website and a stock bars dataset. The study used a rule-based approach to enter trades, with a stop-loss set at 1 average true range, or ATR, which is a measure of volatility. The target was set at 3 ATRs. The study also used a daily open, high, low, and close, or OHLC, proxy to determine entry and exit points. The results showed a return of 1.3255, a drawdown of 10.0409, and a win rate of 41.18. The study used a vectorized portfolio with open and close marks, and 1 basis point fees and 15 basis points of slippage on each side. The return agreed with manual compounding to less than 3e-15. The study found no qualified entry. The historical dates used were actual release dates, but a full pre-announcement audit was not done. This study is a new reserved historical holdout for ConAgra Brands, not prospective forward data.
A study of ConAgra earnings was done from 2021 to 2026. The study used actual earnings data from the ConAgra website and other sources. The data was used to train a trading model from 2021 to 2024 and test it from 2025 to 2026. The model used a simple stop-loss strategy, which is a rule to sell a stock when it falls by a certain amount, in this case, one Average True Range, or ATR, which is a measure of how much a stock's price moves. The model also used a target price, which is three ATRs away from the entry price. The results showed a return of -5.14% and a drawdown of 8.47% over six trades. The model had a win rate of 33.33% and an average win of 3.35%. The study found no qualified entry, meaning the model did not find a good time to buy the stock. The study used historical data and did not validate the results with actual executions. A news scan found the source of the data to be from the ConAgra website and other online sources.
A study on Conagra Brands (CAG) earnings was conducted using actual data from 2021 to 2026. The study removed a duplicate entry from January 4 and used fixed returns before 2025. The data was sourced from the Conagra Brands website and other online sources, including https://www.conagrabrands.com/investor-relations/financial-news/earnings-announcements and https://runway.adityaperswal.workers.dev/data/stock-bars. The study used a rule to enter trades at the open price 10 days before an event, with a stop loss set at 1 times the average true range (ATR) and a target set at 3 times the ATR. The study also used a daily high-low-close-open (OHLC) proxy to adjust for gaps and stops. A vectorized backtest was run with a portfolio that included open and close marks, 1 basis point fees, and 5 basis points slippage on each side. The results showed a return of 4.83%, a drawdown of 8.95%, and a win rate of 41.18% over 17 samples. The study also found that the half Kelly fraction was 3.58 and the manual VBT error was 8.60e-16. The study used a T5 model and found that the T3 and T10 models were sensitive to the data, but no qualified entry was found. The historical dates used in the study were actual release dates, but a full pre-announcement audit was not done. This study is a new reserved historical holdout for CAG and not prospective forward data.
A study of ConAgra Brands earnings was done from 2021 to 2026. The study used actual earnings data from the company's website and other sources. The data was used to train and test a trading strategy. The strategy used a 3-day average true range, or ATR, to set stop-loss and target prices. The strategy was tested on data from 2025 and 2026. The results showed a return of 2.32% and a maximum drawdown of 5.65%. The strategy had a win rate of 50% and an average win of 2.63%. The study also found that the strategy had a high half-Kelly fraction, which is a measure of risk. The study did not validate the execution of the trades and did not audit the pre-announcement data. The results are based on historical data and not prospective forward data. A news scan found the source of the data to be from the company's investor relations website.
A study of ConAgra Brands earnings announcements from 2021 to 2026 found that training a model on data from 2021 to 2024 resulted in a return of -2.9116% with a drawdown of 5.862%. The model used a simple moving average and average true range, a measure of stock price volatility, to set stop-loss and target prices. The model was trained on 17 data points and tested on 6 held-out data points from 2025 to 2026. A 15 basis point slip in price was assumed for each trade, and a 1 basis point fee was charged. The results showed a win rate of 47.06% and an average win of 1.1768% and an average loss of 1.3521%. The model did not select a best holdout, and no qualified entry was found. The study used historical data and did not validate execution or audit pre-announcement data. A news scan found the source of the earnings data at https://www.conagrabrands.com/investor-relations/financial-news/earnings-announcements. The study also used data from https://runway.adityaperswal.workers.dev/data/stock-bars for explicit SIP adjustment. The half-Kelly fraction, a measure of optimal bet size, was 0. The study found that the two-outcome half-Kelly raw value can exceed 1 for variants, but this is not a leveraged authorization, and the qualified primary allocation is 0.
A study of ConAgra earnings was done using data from 2021 to 2026. The study used actual earnings data from the company's website and other sources. The data was used to train a trading model from 2021 to 2024 and test it from 2025 to 2026. The model used a simple rule to buy and sell stocks: it would buy on the third day before an earnings announcement and sell on the day before. The model also used a stop-loss order to limit losses. The results of the study showed a return of 3.55% and a maximum loss of 5.27%. The model had a win rate of 50% and an average win of 2.83%. The study also found that the model had a high sensitivity to certain parameters, but no qualified entry was found. The study used historical data and did not validate the results with actual executions. The data used was from actual earnings releases and was not prospective or forward-looking.