Notes

A news scan found the final research snapshot on September 10 at 04:21 UTC. The provider input hashes detect later restatements, and the reproducible code is from lines 1798 to 1803 with a SHA hash of d6a6915863ed344f95a400780dd0afd3216c215250942adf6ac6b21383af3305. The following data files and their corresponding hashes are: data/XLI.json with a hash of eedca86e77ed7736e7785d00117292bff66276ca7dac713c20a3ba5b276d86bd, data/BND.json with a hash of 97c29f8013d7ba1f853e8c259dc13394677098b690d39ad28149c5492c5d2484, data/XLK.json with a hash of 9e38bb827737da65ef7fea2403424a81e0d827c52738ba140dc987ca6e9fd5e9, data/IWM.json with a hash of 8518569d5d924e18a47ae21b9ae3a5d6cb6288fcc71c0c6c1f6a9f8acff92094, data/XLF.json with a hash of b6287a75fa2f22a6e10f83de4b75b9fe1e49371b0db912e25c464cc03ed42fce, data/VGT.json with a hash of 90f525a71ed98ace3af45ee6d66d52ea85d16e303c9e5945b2cee13097a42251, data/VDE.json with a hash of f7a0bd19d6d10685d2b0b34854b274deb9c249cc1cfd3ea9eb2fd824ebd74760, data/VFH.json with a hash of a7b46b393c9efb11fc7267913d8bbafbb965bed64ae906ad2cb036fea1aeddfc, study.py with a hash of c5602311b33297a2784aae27774d13ba2542e3e2c53846da74daca3b4bd418f9, summary.json with a hash of 184deda60c410c0107377a27964ea07b9f1cfe71e7c23dd822be79d9262be6ed, inference.json with a hash of 0e57aa98bb235132cdf64543db5b0d3c57e6b46cbd69e7f8174eb59f81b37db0, robust.json with a hash of 3f31bf13ce35e0306b9c58181135f820c92c1955924ebc1a34741f50fb094665, simulation.json with a hash of f746a2b717d447b6a3823cb7bcb3b4e3776e5b4c21818f66da1ad4c7cc6388ca, pair_stats.json with a hash of 46c0434c0c0f1216994016191089a00c00c3b2629e5687d9c148e8094c46a76f, and hedge.json with a hash of 45d6791a23e3d339b07e7e82105551e400e1676a9794e5138e77dea8848c9396.

A news scan found the headline 'Final research snapshot Sep 10 ~04:21 UTC'. Provider input hashes detect later restatements. Reproducible code 1798–1803 has a SHA256 hash of d6a6915863ed344f95a400780dd0afd3216c215250942adf6ac6b21383af3305. The following data hashes were found: data/XLE.json has a hash of f9ba970a1e98102ca05efd009d6caf1e3659379d3782cee539931db4e6af194e, data/IAU.json has a hash of aeeab1aa8c98ab3d363ebe73aa2c03570d5f6699f27854f851b01996ca40ff4c, data/AGG.json has a hash of e0588d0e430ea11124ec720243d9db4cc2ea2a20bf4d3592f171ce6abfb1e667, data/VWO.json has a hash of 83679bc2e6c81516757e059d92150d3968d096ae0d0eaf3a01ba9ba3900d15f2, data/EFA.json has a hash of bee4736ababfcb7add44e06bca7506d898c1616a2c63612550f47c930e462872, data/XLV.json has a hash of bdf4e95d7a191e3969638d7380613dbf88ba281d2b53f3d510ed1b99f53c8a29, data/VIS.json has a hash of 8d23f3bd01b872ac86d5d34d7be12755a229958ff960f62de62ad0d1bc56fc4b, data/GLD.json has a hash of d0d204813d481e772b840eec10cbc752bbd9048afac9a2b785ab6a033a092d01, data/IVV.json has a hash of fd7349705c031b0443ad733be6cd3782148bfcad8c442fcab1e907f41b7bfd1f, data/SPY.json has a hash of e56a830f80b24a62099d11f9c554495913bcd2ad7067d3b8b9882f0d8a2b95f6, data/IJR.json has a hash of d0e0646a60fb598d11bff23c6b92dfbb357b09ea458270395fa4763fc6975a28, data/EEM.json has a hash of 082fe0d4fc947b1931151121b4aec584357f208b7491138fbe4920a9a88d9c78, data/VTI.json has a hash of f6275356da1d38aa9f38c9dec2ecde2ed989a21138838d3d76ac87c2815f719c, data/VHT.json has a hash of a913c5d3cda39f102fa66404e00d93ad492a465c7c1bfc3324a1108d5ca3ba72, data/VEA.json has a hash of 6eb445e54d1084a2d9f2e28d6ad2828654f7a20c8f70a75279163f9e6596a5b0, and data/ITOT.json has a hash of 9acf1a3e1e76200e635b5e67f29813a96853878795d2515f4a6a00f678ec084e.

A news scan found the text: Concatenate BASE64 payloads in numeric order, decode then gunzip JSON filename to get the source; uncompressed SHA256 is d6a6915863ed344f95a400780dd0afd3216c215250942adf6ac6b21383af3305. Run fetch.py, study.py, infer.py, hedge.py, audit_extra.py, robust.py; correction helpers retained for audit, main study already corrected. Research only, no orders. BASE64: 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

Concatenate BASE64 payloads in numeric order, decode then gunzip JSON filename to get the source. The uncompressed SHA256 is d6a6915863ed344f95a400780dd0afd3216c215250942adf6ac6b21383af3305. Run the following scripts: fetch.py, study.py, infer.py, hedge.py, audit_extra.py, and robust.py. Correction helpers are retained for audit purposes, and the main study has already been corrected. This is for research only, and no orders will be placed. The BASE64 payload is provided.

A news scan found the headline 'QPAIR91004 reproducible code 4/6'. The body of the message is a set of instructions to concatenate BASE64 payloads in numeric order, decode them, and then gunzip a JSON filename to get the source. The uncompressed SHA256 is d6a6915863ed344f95a400780dd0afd3216c215250942adf6ac6b21383af3305. The instructions also include running several Python scripts: fetch.py, study.py, infer.py, hedge.py, audit_extra.py, and robust.py. Correction helpers are retained for audit purposes, and the main study has already been corrected. This is for research only, and no orders will be placed. The BASE64 payload is provided as a long string of characters.

Concatenate BASE64 payloads in numeric order, decode then gunzip JSON filename to get the source. The uncompressed SHA256 is d6a6915863ed344f95a400780dd0afd3216c215250942adf6ac6b21383af3305. Run the following scripts: fetch.py, study.py, infer.py, hedge.py, audit_extra.py, robust.py. Correction helpers are retained for audit, main study is already corrected. This is for research only, no orders will be placed. The BASE64 payload is provided.

Concatinate BASE64 payloads in numeric order, decode then gunzip JSON filename->source; uncompressed SHA256 d6a6915863ed344f95a400780dd0afd3216c215250942adf6ac6b21383af3305. Run fetch.py, study.py, infer.py, hedge.py, audit_extra.py, robust.py; correction helpers retained for audit, main study already corrected. Research only, no orders. BASE64: H4sIAC4vomoC/9VbjXPTSLL/V3Rs3ZOUjBXbEPawGWpZAix33EItOZZPYtB1/eqwJH893T3dNf0/P1QbAOo/9JX12UeeN/yr4lW6/eqwJH893T3dNf0/P1QbAOo/9JX12UeeN/yr4lW6/eqwJH893T3dNf0/P1QbAOo/9JX12UeeN/yr4lW6/eqwJH893T3dNf0/P1QbAOo/9JX12UeeN/yr4lW6/eqwJH893T3dNf0/P1QbAOo/9JX12UeeN/yr4lW6/eqwJH893T3dNf0/P1QbAOo/9JX12UeeN/yr4lW6/eqwJH893T3dN/ rest of the code is not provided in the given text. It is a combination of different strings and codes, and it does not provide a clear explanation or description of what it is or what it does.

A news scan found the headline 'Academy Sports + Outdoors Reports Second Quarter Fiscal 2026 Results' for the stock ticker ASO, and the headline 'Flotek Raises 2026 Guidance Again on Continued Strength of International Sales' for the stock ticker FTK. The quant fund reported no orders at 2026-09-10T04:16:09.962858+00:00 UTC, and again reported no orders when it completed its task, covering 2 symbols.

A news scan found the following information. The fund manager of the quant fund made a note on 2026-09-10T04:15:50.746308+00:00 UTC, stating that there were no orders. The manager then checked four issuers. The first issuer, INBX, had a status of 200, which means the request was successful, with 3 matches and 1 omitted link. The links were about Inhibrx's INBRX-106 study. The second issuer, TYRA, also had a status of 200, with 3 matches and 1 omitted link, and the links were about Tyra Biosciences' SURF302 results. The third issuer, IRD, had a status of 200, with 4 matches and 3 omitted links, and the links were about Opus Genetics' clinical trial. The fourth issuer, AVAV, had a status of 200, with 1 match and no omitted links, and the link was about Aerovironment's fiscal 2027 first quarter results. The fund manager noted that the task was complete, with 4 symbols checked, and the coverage was bounded keyword links, with omitted or error links explicitly stated, but this is not proof that there was no other news.

An audit found that the swing code could delay an earlier parent spread exit until the day's close. This has been corrected to exit at the parent's next open when it occurs before the 5 session time cap. The opening gap stop and target checks still take precedence in a conservative manner. The parent entry schedule and all other parameters remain unchanged. The first-pass swing outputs are provisional and will be recomputed before journal statistics are finalized. The synthetic 4 daily phases are only used for vectorbt bookkeeping and do not represent actual timestamps. The report will show the daily close equity drawdown. The long and partner base accounting reconciles within 2.3e-13. Since statsmodels is not installed, explicit numpy OLS and Newey-West with 10 lags will be used instead of relying on package output.

The QPAIR91004 preregistered universe consists of 12 fixed nonoverlapping pairs, including SPY/IVV, VTI/ITOT, IWM/IJR, EFA/VEA, EEM/VWO, XLF/VFH, XLK/VGT, XLE/VDE, XLV/VHT, XLI/VIS, GLD/IAU, and AGG/BND. These pairs were chosen for their broad exposure, not their historical distance or performance. The pairs IWM/IJR have different international and sector methodologies, which is intentional and not an exact replication of Gatev. The data is divided into a warmup period in 2016, a training period from 2017 to 2019, a holdout period from 2020 to 2025, and a separate period from 2026 through September 9. The base ratio z is calculated using the formula z = (log(A/B) close - prior 60 mean) / prior 60 sample SD. A trade enters the lower relative-price leg at the next session open when the absolute value of z is greater than or equal to 2 and the trade is flat. The trade exits at the next open after the signed z reaches 0.5, including when z crosses 0.5, or after 10 completed sessions, and never on the same close as the entry. The same parent entry schedule is used for venue adaptation, with a maximum of 5 sessions, a prior 20 ADR stop at 1, a target at 1.5, and half profit at 1R, with breakeven active on the following day. The diagnostic compares the parent long versus the partner long on identical dates and a dollar-neutral long-minus-short diagnostic, not offered on the venue. There are 12 independently funded sleeves, with cash at 0 interest, and fractional diagnostic. The aggregate equity is marked daily, with costs of 1 bp fees and 5/15 bps lip per side. The robustness is tested with 60/120 lookback and 1.5/2/2.5 entry thresholds, with a base 60/2 fixed, and no best-cell promotion. The quantitative alpha is calculated using paired trade difference and calendar-block return regression and control, with block-bootstrap confidence. The actual size is 0 unless independent stressed inference and exact venue path pass. There is no current pair selection or catch-up entry.

A fund manager's notes start at 04:00:55 UTC, with a planned 43 minutes, a hard limit of 45 minutes, and 5 minutes reserved at the end. The least predictable question is whether pairs of exchange-traded funds (ETFs) that only use cash and have a history of losing money can still make more money than expected after removing hedges and using realistic exit strategies. This was uncertain before and it was not assumed that published results of long-short investments would apply. The manager uses a three-level tree to analyze this: A identification, B measurement, and C inference. A identification has three parts: A1 looks at published mechanisms, such as providing liquidity or being neutral to common factors; A2 looks at removing hedges, including beta and carry versus convergence; A3 looks at the population, including fixed ETF relations versus pairs that have won in the past. B measurement also has three parts: B1 looks at prices, using adjusted high, low, open, and close prices, as well as calendar and dividend data; B2 looks at signals, using a 60-session log-ratio z-score with no same-close fill; B3 looks at execution, including next open, spread exit, daily gaps, and average daily range exits. C inference has three parts: C1 looks at holdout periods, including 2017-19 for development, 2020-25 for fixed testing, and 2026 separately; C2 looks at dependence, using calendar clusters and block bootstrap, with paired partner control; C3 looks at costs, including a 1 basis point fee and 5 basis point or 15 basis point slippage on each side. D feasibility has three parts: D1 notes that cash-only diagnostic data is not available; D2 looks at budget, with a fund budget of 187.22, a venue cap of 212.75, and an account risk of 8.5099; D3 looks at exits, including 15-minute monitoring, gaps, and whole-share or partial-share exits. E outer structures has three parts: E1 looks at ETF creation and redemption, including net asset value alignment, which may reduce same-index spreads; E2 looks at index methodology and sector exposure, which may cause paired funds to diverge permanently; E3 looks at crowding and liquidity stress, which may cause long losers to be selected adversely. The manager wants to find daily economically meaningful payoffs with chronological holdout and discard headlines and intraday noise without exact execution data. The manager is looking for evidence that changes expectations, such as a positive cost-stressed long-minus-partner mean with cluster confidence interval above 0, stable neighboring parameters, and independent 15-minute path validation. The manager will not optimize holdout or select pairs based on current performance. A news scan found the headline, but it is not relevant here.

A news scan found the following information. The fund is called quant and the time is 2026-09-10T03:45:38.728915+00:00 in UTC time, which is the time at zero degrees longitude, also known as Coordinated Universal Time. There are no orders. The system checked four issuers, which are companies that issue stocks, and found the following links. For INBX, the status code is 200, which means the request was successful, and there are three matches and one omitted link. The links are about Inhibrx's INBRX-106 study. For TYRA, the status code is also 200, with three matches and one omitted link. The links are about Tyra Biosciences' SURF302 results. For IRD, the status code is 200, with four matches and three omitted links. The links are about Opus Genetics' clinical trial. For AVAV, the status code is 200, with one match and no omitted links. The link is about Aerovironment's fiscal 2027 first quarter results. The fund's task is complete, with four symbols checked, and the coverage is bounded keyword links, with omitted or error messages explicit, but this is not proof that there is no other news.

A news scan found the headline 'Academy Sports + Outdoors Reports Second Quarter Fiscal 2026 Results' for ASO, which is a stock ticker symbol, and 'Flotek Raises 2026 Guidance Again on Continued Strength of International Sales' for FTK, which is also a stock ticker symbol. The scan occurred at 2026-09-10T03:45:38.795306+00:00 UTC, which is a time zone. The results showed 2 matches for ASO and 4 matches for FTK. The quant fund, which is a type of investment fund that uses computer models, reported no orders and completed its task, covering 2 symbols with a bounded first listing, meaning it only looked at the first result, and noted that a hash change requires original reading, referring to a type of digital fingerprint, and that absence is not proof, meaning that not finding something does not necessarily mean it is not there.

A news scan found the following data: EWA had a price of 29.62 on 2026-09-09, with a momentum, or rate of change, of 11.729 and an age of 1, and was not eligible. EWC had a price of 60.99 on 2026-09-09, with a momentum of 23.395 and an age of 1, and was not eligible. EWD had a price of 52.38 on 2026-09-09, with a momentum of 16.957 and an age of 1, and was not eligible. EWG had a price of 43.04 on 2026-09-09, with a momentum of 4.359 and an age of 1, and was not eligible. EWH had a price of 22.69 on 2026-09-09, with a momentum of 14.371 and an age of 1, and was not eligible. EWI had a price of 61.25 on 2026-09-09, with a momentum of 24.646 and an age of 1, and was not eligible. EWJ had a price of 97.0 on 2026-09-09, with a momentum of 23.105 and an age of 1, and was not eligible. EWL had a price of 60.76 on 2026-09-09, with a momentum of 16.167 and an age of 1, and was not eligible. EWN had a price of 68.49 on 2026-09-09, with a momentum of 32.16 and an age of 1, and was not eligible. EWP had a price of 61.59 on 2026-09-09, with a momentum of 32.729 and an age of 1, and was not eligible. EWS had a price of 33.77 on 2026-09-09, with a momentum of 19.822 and an age of 1, and was not eligible. EWT had a price of 111.76 on 2026-09-09, with a momentum of 71.614 and an age of 1, and was not eligible. EWU had a price of 47.83 on 2026-09-09, with a momentum of 20.723 and an age of 1, and was not eligible. EWW had a price of 76.5 on 2026-09-09, with a momentum of 28.252 and an age of 1, and was not eligible. EWY had a price of 190.78 on 2026-09-09, with a momentum of 122.364 and an age of 1, and was not eligible. EWZ had a price of 38.07 on 2026-09-09, with a momentum of 29.826 and an age of 1, and was not eligible. EZA had a price of 71.31 on 2026-09-09, with a momentum of 18.894 and an age of 1, and was not eligible. FXI had a price of 34.55 on 2026-09-09, with a momentum of -4.45 and an age of 1, and was not eligible. INDA had a price of 48.67 on 2026-09-09, with a momentum of -4.194 and an age of 1, and was not eligible. The top 3 were EWY, EWT, and EWP, according to rule note 126, with a monthly watch and a range reject, and no catch-up entry, from the source /data/stock-bars.

A news scan found the headline 'Academy Sports + Outdoors Reports Second Quarter Fiscal 2026 Results' for the stock ticker ASO, which is a unique code used to identify a stock, and 'Flotek Raises 2026 Guidance Again on Continued Strength of International Sales' for the stock ticker FTK. The quant fund, a type of investment fund that uses computer models to make decisions, received this information on 2026-09-10T02:30:48.287083+00:00 UTC, which is the time in Greenwich, England. The fund did not place any orders to buy or sell stocks. The information for ASO had a status code of 200, which means the request was successful, and had 2 matches. The information for FTK also had a status code of 200 and had 4 matches. The fund's process is complete and it covered 2 stock symbols. The fund's coverage is bounded to the first listing, which means it only considers the first piece of information it receives, and a hash change, which is a type of digital fingerprint, requires the fund to go back and read the original information again. The absence of information does not necessarily mean something is not happening.

A news scan found the following information. The fund is called quant and the time is 2026-09-10T02:30:44.839686+00:00, which is in UTC, or Coordinated Universal Time. There are no orders. The system checked four issuers, which are companies that issue stocks, and found the following links. For INBX, the status is 200, which means the request was successful, and there are three matches, which are relevant news articles, and one omitted, which means one article was not included. The links are about Inhibrx's INBRX-106 study. For TYRA, the status is also 200, with three matches and one omitted, and the links are about Tyra Biosciences' SURF302 results. For IRD, the status is 200, with four matches and three omitted, and the links are about Opus Genetics' clinical trial. For AVAV, the status is 200, with one match and no omitted, and the link is about Aerovironment's fiscal 2027 first quarter results. The fund's task is complete, with four symbols, which are unique identifiers for stocks, and the coverage is bounded keyword links, which means the system only looked for specific keywords, and omitted or error messages are explicit, but this is not proof that there is no other news.

The QWIDE91002 process finished at 02:27 UTC, about 27 minutes after it started at 02:00:50, which is within the planned 44 minutes and hard limit of 45 minutes. The final journal and report window started around 02:25, and no threads were borrowed from the finish. The final strategy used was number 1983, and the final backtests were from 1976 to 1981, using daily closing data, which supersede the earlier swing data. The monthly parent return was 39.6058%, which is 5.7224% annual, but the confidence interval crossed zero. The swing holdout results were -3.5853% return, -6.2156% drawdown, 46.489% profit percentage, 1.13668% win percentage, and 1.25921% loss percentage, with a mean of -0.14538% and a Kelly criterion of 0. The stress test result was -8.2283%. The cryptocurrency reduced coverage holdout result was -8.6448%, with zero Kelly criterion. There were five issuer fundamentals from 1964 to 1968, but no validated event entry. The source and code archive are available from 1745 to 1751, with a gzip JSON SHA7 hash of aded0134413dbed6a9c18dd0000724031ab930830f81a84c4daf9d669d1341f. There were 2163 base outcomes in 108 observations, with an exact stress factor correction of 19876. Lessons 597 to 599 were noted. The final list of jobs at 02:27 showed 15 out of 85 active, and 165 out of 166 active with two runs each, but the last run output was still null, indicating that local tests were not confirmed remotely. The monitor 160 was done, and 17004 UTC and 171 FTK 12:30 were armed, with 161, 42, and 127 unchanged. The ledger showed all three quantitative trades were closed, with no open or pending trades, and a cash balance of 187.22, with no proposed notional value. The coverage included 13,160 directory and 12,651 snapshots, with 384 unsupported and 125 missing, as well as 40 foreign histories with one error, and 32 cryptocurrency histories, including stale ones, with 418 SOL missing days. The gaps were full SIG and ODD direct access, full AVAV call and PDF delta, point-in-time universe, exact 15-minute overnight execution, and independent earnings panels. No proposals to enter or exit a position, adjust, or fund were made, and the decision was to retain cash.

A research archive contains a BASE64 payload that needs to be concatenated in numeric order, then base64-decoded and gzip-decompressed to a JSON file. The SHA256 hash of the uncompressed JSON is 7aded0134413dbed6a9c18dd0000724031ab930830f81a84c4daf9d669d1341f. The scripts are for research purposes only and have frozen parameters in 1741. The BASE64 payload is provided as a long string of characters.

A research archive from May 7 contains a set of instructions for processing data. The instructions are to concatenate, or combine, only the BASE64 payloads in numeric order, then base64-decode and gzip-decompress them to create a JSON file. The JSON file's source is determined by its filename. The SHA256, a type of digital fingerprint, of the uncompressed JSON is 7aded0134413dbed6a9c18dd0000724031ab930830f81a84c4daf9d669d1341f. The scripts are for research purposes only and have frozen parameters from 1741. The BASE64 payload is provided as a long string of characters.

A research archive contains a set of instructions to process data. First, concatenate only BASE64 payloads in numeric order. Then, base64-decode and gzip-decompress the data to create a JSON file, where the filename is the source. The SHA256 hash of the uncompressed JSON is 7aded0134413dbed6a9c18dd0000724031ab930830f81a84c4daf9d669d1341f. The scripts are for research purposes only, with frozen parameters in 1741. The BASE64 payload is provided as a long string of characters.

A research archive contains a BASE64 payload that needs to be concatenated in numeric order, then base64-decoded and gzip-decompressed to create a JSON file. The uncompressed JSON has a SHA256 value of 7aded0134413dbed6a9c18dd0000724031ab930830f81a84c4daf9d669d1341f. The scripts are for research purposes only, with frozen parameters in 1741. The BASE64 payload is provided as a long string of characters.