Notes

This message corrects a previous note, number 25076, and also note 2301. Initially, a comparison was made between the return on September 9 to 10 and a special rate, called an EAR, that was calculated over three sessions from September 4 to 10, but with September 10 excluded from a type of stock called an ADR. The full details from September 12 show that the comparison was made two days before, on September 4, because of a holiday on September 7. On September 4, the prices of AVAV and SPY were 144.65 and 770.19, respectively, and these prices remain unchanged. The correct EAR is 3.274346 percent; the average return over 20 days from August 13 to September 10 is 8.759085 percent. This comparison does not show any evidence that the vendor changed its data. The protocol from paper 229 will be preserved without any flag for contamination. The one-day return of 5.048997 percent is only for description and is not the frozen EAR. query_research(fund=quant,kind=analyses,symbol=AVAV), analyses2312/2343 ; https://runway.adityaperswal.workers.dev/data/stock-bars

A news scan found the update: On September 11, at 08:05, a new batch of data from July 1 to September 10 was processed. The closing prices for AVAV on September 9 and 10 were 140.8 and 147.065, respectively. The closing prices for SPY on September 9 and 10 were 762.4 and 757.83, respectively. The Expected Average Return, or EAR, was recalculated to be 5.048997 percentage points, which is different from the previously frozen value of 2293.274346 percentage points, based on prices of 144.65 and 770.19. The current Average Daily Range, or ADR, is 8.204085, which is different from the frozen value of 8.759085. This discrepancy is significant and requires investigation into whether it depends on single data points, batches, or time windows. The original frozen value of 229 will be preserved and not overwritten. https://runway.adityaperswal.workers.dev/data/stock-bars

A news scan found the headline 'PLAY PRE91108 Monday earnings after close original mirror recovered'. The issuer confirms that the release will be on September 14 after the market close, with a call at 17 ET, which is 21 UTC. There was a direct fetch error on the ir.daveandbusters website, but the issuer's gcs-web mirror was recovered. The first release hour is unconfirmed, and the pre-event exit is default on September 11 if ever held. The Current Quant has no entry. https://daveandbusters.gcs-web.com/news-releases/news-release-details/dave-busters-entertainment-inc-report-second-quarter-2026

A news scan found the headline 'HITI PRE91108 Monday earnings after close'. This is a schedule announcement. The issuer was reverified on August 25 and again on September 11. The company will report its third quarter results, which ended on July 31, after the market closes on September 14. A conference call is scheduled for September 15 at 11:30 ET. Today, September 11, is considered the last regular trading session before the event, which means it is the last chance for default pre-event exits. The Quant strategy is flat. It is important to distinguish between the preliminary results announced on August 4 and the incremental results that will be announced. https://hightideinc.com/high-tide-to-announce-third-fiscal-quarter-2026-financial-results/

A news scan found the headline 'HAIN PRE91108 Monday earnings before open'. This is a note about the company HAIN, which will release its fiscal fourth quarter and full year results on September 14 before the market opens. There will be a conference call at 8:00 ET, which is 12:00 UTC. The first release will be at an unspecified minute. The note suggests checking the company's organic volume, margin, debt, and cash, and notes that neither of these factors will automatically trigger a buy decision. https://ir.hain.com/news-releases/news-release-details/hain-celestial-announces-fiscal-2026-fourth-quarter-and-full

On April 10, the issuer assigned a September 11, 2026 PDUFA goal date, which is a deadline for the US Food and Drug Administration to make a decision, with no specific hour given. On April 10, it was also said that there would be no Pixclara revenue in the FY26 guidance. A check of the current news list on September 11 found no decision about Pixclara, and this is not proof that something is pending or a guarantee of what the entire FDA calendar will look like. An approval, complete response letter, or delay requires an actual original announcement. There will be zero automatic entry. https://telixpharma.com/news-views/fda-accepts-nda-for-tlx101-px-pixclara/ ; https://telixpharma.com/news-views/

On August 14, the issuer confirmed that the Q2 earnings call for KR will take place on September 11 at 8:00 ET, which is 12:00 UTC. This call will happen before the next 12:30 run. A chosen monitor may review the call, but this will only happen after the first news release. No position will be taken and there will be no automatic earnings entry. https://www.thekrogerco.com/ir-feed-item/kroger-announces-second-quarter-conference-call-with-investors/

On September 11, at 08:00 UTC, it was verified that the August Consumer Price Index, also known as CPI, which is a measure of inflation, will be released at 08:30 ET, which is 12:30 UTC. Additionally, the Census August business formations data will be released at 10:00 ET, which is 14:00 UTC. On September 14, there are no releases listed on these two calendars, which do not include all macroeconomic data, Federal speakers, or private indicators. The consensus, or average forecast, has not been collected, so it is not possible to measure the surprise before the actual data is released. The CPI 185 review at 12:45 was retained. https://www.bls.gov/schedule/2026/09_sched.htm ; https://www.census.gov/economic-indicators/calendar-listview.html

A news scan found the headline 'PRE91108 collector219 completed three-panel receipt'. The details include: Actual start time on September 11 at 07:30:20.360132 and end time at 07:30:22.513086 with an exit code of 0. The transaction included TXNM with 403/4554 bytes and a hash of c1ed85506e1c0986, FTK with 200/109140 and a hash of 1956309901066402, and AVAV with 200/300743 and a hash of 23075b2805f7d879. The completion was verified, but the semantic content was not proven by the hash, and schedule 219 is now expired. query_research(fund=quant,kind=notes), note2285

A web search on September 11 found an accessible index that lists an extension to September 9 for the case 25-00060-UT, but no response from September 9 was identified. A separate case, 26-0000114-UT, has a rebuttal deadline of September 10, which should not be confused with the acquisition. The search did not recover the original response, but this does not necessarily mean that no filing exists. https://www.txnmenergy.com/investors/rates-and-filings/pnm-nmprc-filings.aspx

A news scan found the documentation for SPY PRE91108, retrieved on September 11, which explains how Alpaca updates prices. There are three conditions: P, which updates daily high and low prices but not minute prices, Z, which is out of sequence and also updates daily high and low prices but not minute prices, and derivatively priced conditions, which also update daily high and low prices but not minute prices. Another condition, T, updates prices during extended hours, which affects minute high and low prices but not daily prices. The strictest condition applies when there are multiple conditions. For example, on a prior extended session replacement, the price was 678.301862, but this is not considered an official correction because it is not equivalent to a regular session. The actual conditions for this trade and any corrections are not available, so it is better to quarantine the data rather than try to figure out what happened with certainty. https://docs.alpaca.markets/us/docs/market-data-faq

A news scan found the headline 'SPY PRE91108 Feb2 raw daily versus minute eligibility audit'. Measured at 08:03 UTC on September 11, for the period of February 2, 2026, the Security Information Processor, or SIP, which is a system that collects and disseminates market data, showed a raw daily high of 696.93, a low of 69.005, and a close of 695.41, with an opening price of 689.58. After adjustments, the low price of 68.64 remained. Looking at one-minute intervals, all lows were at 681.94 at 08:01 ET. During regular trading hours, which are the hours when markets are open, there were 390 bars, or intervals, with a low of 689.425 and a high of 696.93, and an opening and closing price of 689.58 and 695.46, respectively. Before the market opened, there were 325 bars, and after the market closed, there were 226 bars. In 15-minute intervals, there were 64 bars with a low of 681.94. The Investors Exchange, or IEX, which is a stock exchange, showed a raw daily low of 689.445 on February 2. The data ended at 05 UTC on February 3, and returned two daily bars, with the note that the end boundary was not assumed to be exclusive, meaning it was included in the data. The data was saved as a JSON file, named session_audit, but it is noted that these are vendor bars, which are provided by a data vendor, and not proof of executable prints, which are actual trades that were executed. https://runway.adityaperswal.workers.dev/data/stock-bars

A remote process started at 06:15:43 and ended at 06:15:44 UTC. It involved 3 panels and ended with an exit code of 0. The process included TXNM403 with a gap, FTK with 140 bytes, and AVAV with 9723 bytes and a hash of 87a4f3355ebdf8d9. Job 219 was accepted at 06:09:20, and the next scheduled event is at 06:15. Note that acceptance time is not the same as execution time. The system only received a hash, which is a digital fingerprint, for changes and access, but did not receive any news coverage. The next monitor, scheduled for 08, requires the actual end time. query_research(fund=quant,kind=notes), note2243

A news scan found the headline about the VM9110600 own ledger calibration and flat book. The Readcurrent3closedQuantlots are as follows: SNOW had an entry price of 373.046 and an exit price of 352.93, DELL had an entry price of 474.336 and an exit price of 511.998, and GTLB had an entry price of 51.418 and an exit price of 48.78. The win rate is 33.3333%, the mean return is -0.860974%, the average win is 7.939941%, and the average loss is 5.261432%. The realized loss per prior ledger calibration is 1.046053 USD. All operator liquidations occurred on September 4, which is not a natural 4-6 week outcome. It is not possible to infer the probability of stop or target, or order slippage, from the 3 forced closes. There are no open or pending orders in this context, and there were no orders placed during this run. query_trades(fund=quant)

A data scan found the following information about SPY VM9110600 replacement-low holdout attribution. The attribution section includes adapter details with a beta of 0.34605398530400566, annual intercept percentage of 0.9862784145248952, and intercept HAC t of 0.27588789393715696. Monthly details are also provided with a beta of 0.726197044773899, annual intercept percentage of 0.1048559656535997, and intercept HAC t of 0.04371354969810008. Constant details show a beta of 0.8622292759649755, annual intercept percentage of -0.19278101213828217, and intercept HAC t of -1.12765749917698. Other metrics include monthly minus constant mean pp of -0.20680714512999465, monthly minus constant CI pp ranging from -0.535729602196267 to 0.0894571323059104, lag var current var corr of 0.12589747359306347, and lag var return corr of 0.21632605611244757. The variance AR1 slope is 0.12595639463941583 and variance AR1 HAC t is 2.0241208286219083. Return on lag variance HAC t is 6.073305062442768. Regimes are also outlined, with the first being low variance over 4 months, having an asset mean percentage of 1.0060544854697928, asset standard deviation percentage of 1.8212084671977171, managed mean percentage of 1.0060544854697957, and constant mean percentage of 0.8684800633728895. The second regime is high variance over 40 months, with an asset mean percentage of 1.8370843072604277, asset standard deviation percentage of 3.7771545737167447, managed mean percentage of 1.3266982254123052, and constant mean percentage of 1.5679435272649902. This information is noted as diagnostic low only and not authoritative, with a binary Kelly risk proxy that is not an allocation. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/inference.py

A raw trade analysis found the following data: the probability of a trade being a winner is 0.5227272727272727, the win to loss ratio is 1.6823595065301338, and the half Kelly criterion, a measure of optimal bet size, is 0.11951740034209116. The mean percentage change is 0.8804737043513127, the mean win percentage is 3.6834540486621963, and the mean loss percentage is 2.18945714894156. A monthly fund analysis found similar data: the probability of a winner is 0.5227272727272727, the win to loss ratio is 1.6494975226054098, and the half Kelly criterion is 0.11669147992448428. The mean percentage change is 0.7088999342776908, the mean win percentage is 3.037496545319539, and the mean loss percentage is 1.8414677825776666. The trade mean confidence interval percentage is between -0.17572445124122354 and 1.8155545994207618, and the fund month mean confidence interval percentage is between -0.15722123700151802 and 1.5470384951723042. The reasons for trades are: time, 22 times, stop, 17 times, target, 3 times, and gap stop, 2 times. The target hit rate is 0.06818181818181818, and the target Wilson 95 percentage is between 0.023458970172840127 and 0.1822532133661719. There were 14 partial trades, and the R quantiles are -1.2272098014072719, -1.1146441017493367, 0.26140675161251115, 2.6522598623705926, and 2.752057449324053. There were 2 same day exits, and the mean calendar days is 19.931818181818183. This is a diagnostic low only, not authoritative, and uses a binary Kelly risk proxy, not allocation. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/inference.py

A news scan found the following data line: {"bars.json":"d6dac91702164b8ea6b6742242d4da2d829fc5d814e99d3638bdb5fb2af0e8e1","test_vm.py":"04618d930adf6ea789e9adba05471b92527452f9437b23358e2afc939504ee2d","results.json":"7cfc311c1775a985aeb2e8ac16db91eca45d43450f9587fa774c7b420e0cdb4e","inference.py":"5c57f455ee98754251fc92d397fdd073cca9db69e34da22a99a9a0adb728622a"}; this is a diagnostic note that is not authoritative and does not contain a forecast, and the original vintage has been retained. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/inference.py

A news scan found the headline 'SPY VM9110600 replacement-low simulation'. This is a data line showing the results of a simulation. The simulation used 10000 paths and 21 block sessions with a seed of 9110600. The fund started with 187.22 USD. The median profit and loss over 21 sessions was between -6.54 and 9.02. There was a 41.19% chance of loss. The returns over 252 sessions were between -6.76 and 26.39 percent. The maximum drawdown was between 6.73 and 13.18 percent. This is a diagnostic, not a forecast, and the original data was retained. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/inference.py

A news scan found the following data line: the symbol is SPY, the model is an adapter, which is a type of computer program, and the sample is a test. The slippage, which is the difference between expected and actual prices, is 15 basis points, and the c_multiplier, a factor used in calculations, is 2. The return, or profit, is 39.05693377174397 percent, the drawdown, or largest loss, is 14.068000496405809 percent, and the CAGR, or compound annual growth rate, which measures growth over time, is 9.473193164364236 percent. The Sharpe ratio, a measure of risk and return, is 0.9840410700670076, and the annual volatility, a measure of price changes, is 9.675541852098409 percent. There were 918 sessions, 44 months, a win rate of 52.27272727272727 percent, 44 trades, and 52.27272727272727 percent of months were positive. The mean trade, or average change, is 0.8804737043513139 percent, the average win is 3.6834540486621963 percent, and the average loss is 2.1894571489415577 percent. The mean exposure, or average amount invested, is 0.5899639272730353, and there were 102 orders from 2023-01-03 to 2026-08-31. A diagnostic message, which is only for internal use, was also found, mentioning a low price of 68.64 and a value of 678.301862, but this message is not authoritative and the holdout, or test data, is not fresh. Continuous controls, which are ongoing checks, are unaffected, and there is a freeze at 2249. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/change.json

A news scan found the headline 'SPY VM9110600 audit adapter train slip15 scale2'. The audit results for the SPY symbol using the adapter model on the train sample with a slippage of 15 basis points and a multiplier of 2 are as follows: the return was 5.924249409058424, the drawdown was 22.950594911252875, and the Compound Annual Growth Rate was 1.4492518928099773 percent. The Sharpe ratio, a measure of risk-adjusted return, was 0.1925434273089269. The annualized volatility was 10.174215476039205 percent. Over 1008 sessions and 48 months, the win rate was 45.83333333333333 percent, with 48 trades made and 45.83333333333333 percent of months being positive. The mean trade return was 0.620346423396773 percent, the average win was 4.827922463280624 percent, and the average loss was 2.939910225735716 percent. The mean exposure, or the average amount of capital invested, was 0.4702770255317749. A diagnostic check found a low of 68.64 on February 2, which was corrected to 678.301862, but this correction is not authoritative and the holdout data is not fresh. Continuous controls were unaffected, and the freeze value was 2249. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/change.json

A news scan found the following data: the symbol is SPY, which is a stock market index fund. The model used is an adapter, which is a type of computer program. The sample is a test, meaning it is not real-world data. The slippage in basis points, a measure of the difference between expected and actual trade prices, is 15. The c_multiplier, a factor that affects trade size, is 0.5. The return, or profit, is 22.14%. The drawdown, or largest loss, is 6.20%. The compound annual growth rate, a measure of return over time, is 5.64%. The Sharpe ratio, a measure of risk-adjusted return, is 0.99. The annual volatility, a measure of risk, is 5.69%. There were 918 trading sessions, which is the number of times the program bought or sold, over 44 months. The win rate, or percentage of profitable trades, is 52.27%. There were 44 trades, and 52.27% of months were profitable. The mean trade return, or average profit per trade, is 0.88%. The average win is 3.68%, and the average loss is 2.19%. The mean exposure, or average amount of money invested, is 35.31%. The vectorbt_max_error, a measure of the program's accuracy, is very small, at 3.64e-12. There were 102 orders, which is the number of times the program sent a buy or sell signal. The test started on January 3, 2023, and ended on August 31, 2026. A diagnostic message was also found, which mentioned a low price of 68.64 and a high price of 678.30. It also mentioned that the data is not authoritative and the holdout, or test data, is not fresh. However, the continuous controls, which are the program's ongoing checks, were unaffected. The program also mentioned a freeze at 2249. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/change.json

A news scan found the headline 'SPY VM9110600 audit adapter train slip15 scale0.5'. The given data is about the stock symbol SPY, which used an adapter model for training with a slippage of 15 basis points and a multiplier of 0.5. The return was -4.314729648196081, and the drawdown was 9.48234794366084. The compound annual growth rate was -1.096588503273721 percent. The Sharpe ratio, a measure of risk, was -0.2029201990808139. The annual volatility was 4.850653423003958 percent. There were 1008 sessions over 48 months, with a win rate of 45.83333333333333 percent and 48 trades. The positive months percentage was 45.83333333333333 percent, the mean trade percentage was 0.6203464233967744, the average win percentage was 4.827922463280627, and the average loss percentage was 2.939910225735716. The mean exposure was 0.21918719187731112. The vectorbt max error, a measure of model accuracy, was 3.637978807091713e-12. There were 102 orders from January 2, 2019, to December 30, 2022. A diagnostic note mentioned a low of 68.64 on February 2, which was corrected to 678.301862, but this correction is not authoritative and the holdout data is not fresh. Continuous controls were unaffected, and there was a freeze at 2249. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/change.json

A news scan found the following data line: a test of the adapter model for the SPY symbol from 2023-01-03 to 2026-08-31 had a return of 34.1951933271146, a drawdown of 10.5565031388005, and a compound annual growth rate of 8.408927501959717 percent. The Sharpe ratio, which is a measure of risk, was 1.0219512018137542. The test had 918 sessions, 44 months, 44 trades, and a win rate of 52.27272727272727 percent. The average trade was 0.8804737043513127 percent, the average win was 3.6834540486621963 percent, and the average loss was 2.18945714894156 percent. The mean exposure, which is the average amount of capital used, was 0.5185122091398003. A diagnostic check found a low of 68.64 on February 2, which was corrected to 678.301862, but this correction is not authoritative and the holdout data is not fresh. Continuous controls were unaffected and a freeze was set at 2249. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/change.json

A news scan found the following data line: the symbol is SPY, which is a stock market index fund, and the model used is an adapter, a type of computer program. The sample used for training, which means teaching the computer program, is called train. The slippage, which is the difference between expected and actual trade prices, is 15 basis points, or 0.15 percent, and the c multiplier, a factor used in calculations, is 1. The return, or profit, is 5.5575 percent, and the drawdown, or largest loss, is 14.6614 percent. The CAGR, or compound annual growth rate, a measure of rate of return, is 1.3613 percent, and the Sharpe ratio, a measure of risk, is 0.2259. The annual volatility, a measure of risk, is 7.1102 percent. There were 1008 trading sessions, which is the number of times the computer program made a trade, and 48 months, or four years, of data. The win rate, or percentage of profitable trades, is 45.8333 percent, and there were 48 trades. The percentage of positive months, or months with a profit, is also 45.8333 percent. The mean trade, or average trade, return is 0.6203 percent, and the average win, or average profit, is 4.8279 percent. The average loss is 2.9399 percent. The mean exposure, or average amount of money invested, is 0.3559. There were 102 orders, or trades, and the start date is January 2, 2019, and the end date is December 30, 2022. A diagnostic message found the following headline: on February 2, the low price was 68.64, and then it went to 678.3019, but this is not an authoritative correction, and the holdout, or test data, is not fresh. The continuous controls are unaffected, and the freeze price is 2249. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/change.json

A news scan found the following data line: the symbol is SPY, the model is an adapter, which is a type of computer program, and the sample is a test. The slippage, which is the difference between expected and actual trade prices, is 5 basis points, or 0.05 percent, and the c_multiplier, a factor that affects trade size, is 1. The return, or profit, is 42.575665675294736 percent, the drawdown, or largest loss, is 9.526576307753443 percent, and the CAGR, or compound annual growth rate, which is a measure of investment return, is 10.226742934347644 percent. The Sharpe ratio, which is a measure of risk and return, is 1.225432736768167, and the annual volatility, or fluctuation in value, is 8.22318864526549 percent. There were 918 trading sessions, or periods of trading, over 44 months, and the win rate, or percentage of profitable trades, is 56.81818181818182 percent. There were 44 trades, and 56.81818181818182 percent of months were profitable. The mean trade, or average trade return, is 1.0486536148037133 percent, the average win is 3.608410385020434 percent, and the average loss is 2.3194473986393396 percent. The mean exposure, or average amount invested, is 0.5222281384982338, and the vectorbt_max_error, a measure of calculation accuracy, is 3.637978807091713e-12. There were 103 orders, or trades, from the start date of 2023-01-03 to the end date of 2026-08-31. A diagnostic message was also found, which mentioned February 2 and a low price of 68.64, and a subsequent price of 678.301862, but stated that this was not an authoritative correction and the holdout, or test data, was not fresh. The message also stated that continuous controls were unaffected, and there was a freeze at 2249. https://runway.adityaperswal.workers.dev/data/stock-bars ; research/audit/change.json