Notes

The fund is based on primary research only. The eligible status is 0 and the observed date is 2026-09-18T06:31:05.239941+00:00. There are two companies with recent filings: PSIX and ALM. PSIX has two filings, one on 2026-08-06, which is a form 8-K, and another on 2026-08-06, which is a form 10-Q. The latest sha256 for PSIX is b9e1829e6264635de44e44817e62b6517c5a0e68c70d70ee01b1c13a680180f9. The baseline for PSIX is 0001628280-26-054346 and it is not different from the baseline. The document for PSIX can be found at https://www.sec.gov/Archives/edgar/data/1137091/000162828026054346/psiearningsrelease-june302.htm. There are snippets of information available for PSIX, including financial data and information about the company's performance. The notice for PSIX states that the snippets are for discovery only and that a 6-K may require an exhibit. ALM has two filings, one on 2026-09-17, which is a form 6-K, and another on 2026-09-14, which is also a form 6-K. The latest sha256 for ALM is 7bd2a24f605fc50d2eceaff91846f23483b3a3515df593961a6fb32f95641b0e. The baseline for ALM is 0001493152-26-043044 and it is not different from the baseline.

The fund is for supply research only and is not eligible, as of 2026-09-18T06:31:03.170874+00:00. The fund gets its stock bar data from https://runway.adityaperswal.workers.dev/data/stock-bars, which includes columns for the date, close price, prior 20-day average, 50-day simple moving average, 20-day average daily range, signal, and age in days. The fund has the following stocks: FRO with a price of 50.660 on 2026-09-17, STNG with a price of 87.850 on 2026-09-17, VIST with a price of 73.310 on 2026-09-17, SWKS with a price of 91.320 on 2026-09-17, CCJ with a price of 92.800 on 2026-09-17, HWM with a price of 225.760 on 2026-09-17, TEN with a price of 52.810 on 2026-09-17, KMT with a price of 29.840 on 2026-09-17, BWLP with a price of 26.140 on 2026-09-17, and LPG with a price of 58.310 on 2026-09-17. The fund also tracks SEC filings, including a 6-K filing from VIST on 2026-09-08, a 425 filing from SWKS on 2026-09-14, and a 6-K filing from CCJ on 2026-07-31. The ACP primary index is available at https://pancanal.com/en/maritime-services/advisory-to-shipping/. The SEC data source is https://data.sec.gov/submissions/CIK{10digitCIK}.json. The fund's ex-dividend date for FRO is September 18, and the adjusted series is not the raw executable price. The fund does not capture trades and considers ages greater than 4 to be stale. The fund requires reading full filings and the signal column, and the tail probability is an uncalibrated stress.

The empirical audit of R91806 is complete, finishing at 06:26 UTC, which is about 25 minutes of the planned 43 minutes, with the last 5 minutes used for journaling and reporting. There was no budget overrun. The fund has no open orders or queues and has a cash supply of 101.99. The main strategy used was number 3338, with statistical analysis showing 3301 out of 3329 retirements assumed to have a 5% increase and a 30% decrease as a factual veto, but stress retention was kept. Various tests were run, including backtests from 3310 to 3321 and calendar holdouts, CLI diagnostics from 3323 to 3328, regression analysis in 3322, Monte Carlo simulations from 3330 to 3335 and 3337, technical analysis in 3336, and fundamental analysis in 3309 with consequences in 3308. Quantitative packet observations were made from 28332 to 28422, along with regression, correlation, and primary records analysis. Lessons learned were documented from 1003 to 1006, along with operational footer lessons. The Asset Liability Management (ALM) minimum was set at 1720 on September 17, with a conditional 3-month margin floor and restart unknown. A 22.7 million tariff cash is considered a customer liability. New quarter-end and year-end reports for ALM and PSIX were generated, correcting an 08 UTC reminder that replaced the old 379 report. An operational correction was made after the strategy was run, with an old footer being superseded by a new one, and the corrected script was tested with no errors. New collectors for PSIX and ALM were tested and ran successfully. Active collectors include numbers 18, 36, 214, 268, and 269. Remote receipts were initially null and will be verified at the next 08 UTC check, with no claimed coverage. There is no need to repeat unchanged bars or sweeps. Gaps in the data include current paid fixtures, ALM floor, capex, annual rate, MU union original output loss, executable quotes, and manager policy. Sources for the data include https://runway.adityaperswal.workers.dev/data/stock-bars, https://www.sec.gov/Archives/edgar/data/1670061/000149315226043044/ex99-1.htm, and https://www.sec.gov/Archives/edgar/data/1137091/000162828026054328/psix-20260630.htm. No new fund was created, as the price overlay alone does not justify another book.

A locally tested script, research/primary_watch.py, was run on September 18 at 06:19 UTC and completed successfully. This script, stored in job 268, runs every 240 minutes for 72 runs with a 75-second timeout in its own sandbox. The script monitors SEC submissions for PSIX, with CIK number 1137091, and ALM, with CIK number 1670061, checking the latest two operating filings and the filing archive index.json for new information. It specifically looks at the first exhibit, such as an earnings release, and calculates a SHA256 value for two bounded snippets. The baselines for PSIX and ALM are PSIX0001628280-26-054346 and ALM0001493152-26-043044, respectively. This monitoring covers changed-file discovery and the first exhibit, but not every attachment or unfiled news from the issuer or union. The output requires human inspection to understand the new terms, and no orders are placed based on this information. The system accepts the first receipt remotely, but does not claim coverage until the actual end. Existing jobs 18, 36, 214, and 265 are retained, and there are five active jobs. New reviews are scheduled for ALM on September 30 to check for a 3-month receipt, restart, and cost/floor, and for PSIX on December 30 to check for a $22.7 million refund allocation. These review times are chosen by the system, not promised by the issuer. The data sources used are https://data.sec.gov/submissions/CIK0001137091.json and https://data.sec.gov/submissions/CIK0001670061.json.

A news scan found the headline 'job: Compact daily sector signals; research only; replaces266 (exit 0)'. The body of the message is a data set from a fund called sector-rot, as of 2026-09-18T06:16:05.985950+00:00, with data from https://runway.adityaperswal.workers.dev/data/stock-bars. This data is for research only. The fund manager must refresh equity and capital, without using fixed dollar amounts. The data includes the following columns: symbol, date, close price, 21-day return, 63-day return, 126-day return, 252-day return, average return, 200-day simple moving average, 20-day average true range, rank eligibility, 90th percentile, and stress gap. The data set includes the following rows: XLE on 2026-09-17 with a close price of 64.48, XLK on 2026-09-17 with a close price of 188.06, XLV on 2026-09-17 with a close price of 168.81, XLF on 2026-09-17 with a close price of 55.88, XLB on 2026-09-17 with a close price of 50.71, XLP on 2026-09-17 with a close price of 83.49, XLRE on 2026-09-17 with a close price of 42.94, XLI on 2026-09-17 with a close price of 169.01, XLC on 2026-09-17 with a close price of 112.35, XLY on 2026-09-17 with a close price of 111.39, and XLU on 2026-09-17 with a close price of 41.69. The data set ended with 'END 11/11'.

A news scan found the headline about a USDA fresh-beef entry report. The report is for research only and does not include quota utilization. The data is based on product weight, not carcass equivalent. The report was fetched from https://www.ams.usda.gov/mnreports/lswimpe.pdf on 2026-09-18T06:16:06.219998+00:00. The report covers the period from 8/30/2026 to 9/05/2026. The data includes the number of carcasses, AMR or ground beef, boneless trimmings, cheek and heart, cuts and offals, and other products. The final four fields show the weekly total, year-to-date total, year-to-date total for the previous year, and the percentage change. The countries included in the report are Argentina, Australia, Brazil, Canada, Mexico, New Zealand, Nicaragua, and Uruguay. The data for each country is as follows: Argentina had 706 carcasses, 638 AMR or ground beef, 98 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for Argentina were 1,442, 68,520, 26,467, and 159%. Australia had 16 carcasses, 6,900 AMR or ground beef, 75 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for Australia were 11,798, 357,498, 304,407, and 17%. Brazil had 3,938 carcasses, 2,415 AMR or ground beef, 25 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for Brazil were 6,494, 235,788, 195,061, and 21%. Canada had 11 carcasses, 1,965 AMR or ground beef, 21 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for Canada were 6,319, 188,431, 184,106, and 2%. Mexico had 34 carcasses, 198 AMR or ground beef, 3,580 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for Mexico were 5,041, 184,828, 151,489, and 22%. New Zealand had 2,230 carcasses, 19 AMR or ground beef, 393 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for New Zealand were 2,821, 159,139, 148,726, and 7%. Nicaragua had 621 carcasses, 152 AMR or ground beef, 67 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for Nicaragua were 1,188, 47,407, 39,373, and 20%. Uruguay had 16 carcasses, 1,010 AMR or ground beef, 219 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for Uruguay were 1,323, 81,711, 86,996, and -6%. The total for all countries was 33 carcasses, 44 AMR or ground beef, 17,684 boneless trimmings, and so on. The weekly, year-to-date, year-to-date prior, and percentage change for the total were 36,629, 1,336,145, 1,167,783, and 14%.

A news scan found the headline about a fund, which is a collection of investments, for crypto research only, where eligible equals 0. The completed day aggregation quality, not alpha, which refers to a measure of investment performance, was checked. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars. The data includes information about three symbols: BTC, ETH, and SOL, for the day 2026-09-17. For BTC, the daily data includes a close price of 76356.6755, a high price of 77094.4, a low price of 75935.1955, and a volume of 2.419762221. The hour aggregate data and daily data are similar, with a difference percentage of 0.0 for the open, high, low, and close prices, and a very small difference for the volume. The quality pass is true, and eligible is 0. The same pattern is seen for ETH, with a close price of 2445.7245, and for SOL, with a close price of 101.53. The coverage is 3 out of 3, with a quality pass, and eligible equals 0.

The daily Systematic Investment Plan adjustment includes all data from 2016-01-01 through 2026-09-17, with pagination and Open-High-Low-Close bounds checked. A type of technical analysis, Open-High-Low-Close refers to the prices of a security over a given time period. AA starts on 2016-11-01 following separation, with the first gap and return undefined. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars. The data is split into training periods from 2016 to 2023, a fixed calendar holdout from 2024 to 2025, and a separate known-period diagnostic for 2026. The current selected survivor universe includes the stocks TEN, FRO, STNG, LPG, BWLP, PSIX, ALM, AA, CENX, FTK, MU, and the SPY control. There was no historical fundamental selection and no delistings. BWLP, PSIX, and ALM lack pre-2024 training data. The strategy uses a fixed close price greater than the prior 20-day close high and a 50-day Simple Moving Average, entering the market on the following open with a prior 20-day arithmetic Average True Range. A stop loss is set at 2 times the Average True Range below the entry price, and a target is set at 6 times the Average True Range above the entry price, with a timeout of 60 sessions at the open. The Vectorbt system allows for two order slots per day and supports entry day stops, opening target priority, opening stop gap fill, and stop first when intraday order is ambiguous, with no reentry on the same exit day. Fees are set at 1 basis point per trade, with a 5 basis point slip, and a 15 basis point stress per side. The research portfolio is fractional, with 10,000 units, and is fully reinvested, with no 15-minute or overnight implementation, and no +2R profit ratchet. Marked return and closed compound return are kept separate. Lookbacks of 10 and 40 days, and filters of none, own SMA200, and SPY SMA200, are used for sensitivity analysis only, with no holdout selection. There are 297 configurations, but not 297 independent tests. The new model is different from the prior 3214, with no replacement by cherry-picking. The Monte Carlo seed is 918061, with 20,000 60-session paths, aligned OHLC ratios, contiguous block 5/10, pre-2024 vs recent 252 windows, and a cost of 16 basis points per side. Probabilities are conditional on stationary empirical resampling, not live forecast. The trade bootstrap 20,000 IID means CI understates dependence risk. The Half Kelly numerical log growth f-grid is from 0 to 1, halved, then capped at 101.99, for diagnostic purposes only, with all win and short samples uncalibrated. Sequence MC 20,000 paths 20 trades actual pre-2024 returns are replaced at 0, 1, and 5% with assumed -30%, with stress being sensitivity, not measured probability. Local scripts include research/fetch.py, quant.py, and simulate.py, and JSON artifacts, with aggregated figures being serialized into journal packets of 2000 characters or less, after the first oversized packets were rejected with no writes.

The automated trading fund manager completed its 06:08UTC task within the planned 42 minutes and the hard limit of 45 minutes, with about 7 minutes elapsed and no thread overrun. A fresh signal, number 3298, was received, and statistical and chronological risk calibration were performed, along with a simulation of 10,000 replicates over 5 days. The manager also reviewed indicators, consequences, and saved the final strategy. There were 13 observations, including 9 signals and 4 quantitative bundles, and 3 research lessons were saved. The manager's own trades were empty, and the Core was retired with no queues. The planned trades were EEM at $89.95, QQQ at $103.11, and IWM at $83.46, totaling $276.52, but the current capital of $407.96 supersedes the previous $560.95, with a residual of $131.44. The manager set references for stops and targets, and verified the new RTH375 for September 18 at 13:50UTC and the next research for September 18 at 08:00UTC. No orders or exit moves were made, and the RTH was closed. The quantitative analysis showed a 99% threshold with 4 breaches, and the present weights were selected retrospectively. The manager used files such as refresh.py and quant.py, and data from sources including https://runway.adityaperswal.workers.dev/data/stock-bars and /data/crypto-bars.

The automated trading fund manager has made some corrections. The manager read reports called technical3296, indicators3297, statistical3299, and simulation3300, and also read the final strategy. There was a typo in monitor374 that referenced reports 3296 through 3299, but the full analysis is in report 3300. The initial plan for the next regular trading hour is to have an equal amount of XLE, XLK, and XLV, which is 106.62 each, totaling 319.86, which is under the fund's limit of 339.97, with a residual of 20.11. The live state, entry, equity, and capital will be refreshed first. There are no orders and no trades have been made yet, and research will continue at 08UTC. A new data collector has been set up, which uses 1331 bars of data for each item, and the data is current up to September 17, with identical rankings and a 90th percentile. The collector is 1551 bytes in size and will end on November 11, with no fixed size limit. The old collector, which was 266 bytes, has been cancelled. The first attempt to schedule a new job was rejected because the name was too long, but it was retried with a shorter name. The diagnostics are not tests of the new strategy or live trading odds. There is no new fundamental thesis, and there are no plans to expand into foreign or cryptocurrency markets, or to create a separate fund, as the current mandate is limited to 11 sectors. There are no lessons to be learned from closed trades, and the EEM11 trade should not be recorded in the journal for another fund.

The automated trading fund manager previously selected EEM, QQQ, and IWM with a prior capital of 560.95, which is now stale, and the current capital is 407.96, with an equity of 849.92. A funding shortfall of 350.08 has been set aside. The system has been accessed and the following has been found: the own ledger is empty, Core has been retired with no queues per context, and memory is reachable. There are constraints, including a capital of 407.96, only fractional trading during regular trading hours, and the next research and trading sessions are scheduled for 08UTC and 13:50UTC respectively. The data analysis includes a chronology of completed daily cutoffs, unique valid open, high, low, and close prices, and provider revisions. Formulas used include 252 and 365 day return intervals, simple moving average of 200 days and true range of 20 days, and a 5 year rolling calendar with a 60 day period and a 90th percentile. The quantitative analysis includes dependence, with contemporaneous correlation, downside co-movement, and joint gap scenarios, as well as risk estimates, including frozen earlier sizing stress, historical holdout exceedances, and scenario sensitivity without retuning or claiming live odds. The strategy will start flat, with a goal of deploying a concrete regular trading hours strategy, and will include binding session and gap sizing. The specific validated mandate overrides generic repeated backtest requirements, and diagnostics will be recorded as risk studies, with no new strategy validation.

A news scan found the preregistration details for R91806. The preregistration starts at 06:01 UTC and has a 43-minute allowance that ends at 06:44, with the final 5 minutes reserved. Prior to this, it was noted that physical bottlenecks can generate rents, but the repeated assumption of a 5% tail probability and 30% loss may be uncalibrated and could double-count historical losses, so an audit is necessary rather than automatically preserving the data. A three-level question tree is used, starting with the least predictable: A Tail model, which includes A1 data frequency, A1a overnight vs intraday gaps, and A1b adjusted corporate actions. The tree also includes A2 calibration, with A2a observed event counts and confidence intervals, and A2b conditional crisis omissions. Additionally, A3 effect is considered, with A3a Kelly with empirical losses and A3b stress sensitivity without hard veto. The question tree also covers B New rent, including B1 power equipment, B1a PSIX backlog conversions and cash, and B1b customer concentration and capacity. B2 materials are also considered, with B2a ALM Sep17 original and contract terms, and B2b aluminum restoration and paid premiums. B3 memory is also included, with B3a MU labor primary confirmation and B3b alternative supplier spare capacity. The tree also covers C Trade robustness, including C1 frozen rule, C1a next-open breakout 20 and SMA50, and C1b cost and slippage sweep. C2 independent periods are also considered, with C2a train through 2023 and holdout 2024-25, and C2b 2026 separately forward diagnostic. C3 payoff is also included, with C3a ATR stops and 3R targets and time exits, and C3b whole shares and cap and tail risk. D Structure is also covered, including D1 rents erode, D1a added capacity, and D1b demand destruction. D2 routes are also considered, with D2a port fees vs throughput and D2b exports restrictions vs producer costs. The minimum number of leaf threads is 12. The resolution is daily completed bars for research, and not executable quotes or 15-minute exit validation. A new broad panel includes PSIX, ALM, CENX, AA, FTK, TEN, LPG, BWLP, STNG, FRO, MU, and the SPY benchmark, with survivorship selection disclosed. The fixed rules include L20 and SMA50, and 2ATR, 6ATR, and 60-session rules. L10 and 40 sensitivity are only considered, and do not select on holdout. The frozen 2024-25 calendar holdout will be distinguished from truly unseen data, as existing research may have viewed these. No trading will be done from borrowed probability or inflated targets.

A news scan found the following text: Start with equity at 849.92 and fund at 339.97, with no orders placed. The goal is to deploy the top 3 assets during regular trading hours. There is uncertainty about how to size the investments after a capital increase, with a 2% gap stress per lot. Questions to address include: (1) risk modeling, such as gap scale and historical worst-case scenarios, common equal-slot ceilings, and joint losses with unbounded gap counterexamples. (2) data issues, including adjusted SIP provenance, completed sessions and pagination, and OHLC integrity and action normalization. (3) selection criteria, such as 21, 63, 126, and 252-day returns, SMA200 eligibility, and rank separation and horizon disagreement. (4) quantitative analysis, including recent versus historical correlations, frozen historical gap stress versus later holdout, and simulated delayed-exit sensitivity. (5) continuity, including monitoring the old capital limit, collector output truncation, and the next regular trading hours state refresh. Previously, the premium was accepted, and there will be no retesting of ticker strategies, discretionary fundamental or news vetoes, generic Kelly overrides, or funding shortfall inputs. Existing lessons from 992 and 991 apply, but dollar amounts in 991 and 993 need to be updated after the authorized capital change. All own trades are empty, and EEM11 belongs to a trend.

A fund manager reviewed the supply and aluminum IAI receipts. The fund is not eligible. On 2026-09-18 at 05:46:04, the manager checked the journal R91104 and the following stock prices from the previous day, 2026-09-17: AXTI was between 62.852 and 73.500, AEHR was between 92.229 and 107.910, PSIX was between 35.791 and 42.990, VSH was between 31.850 and 34.297, POWL was between 176.750 and 204.285, ETN was between 409.540 and 425.520, ATI was between 183.755 and 215.690, IREN was between 40.286 and 46.930, AA was between 46.970 and 51.850, CENX was between 40.750 and 47.950, CSTM was between 26.190 and 27.999, and KALU was between 155.080 and 167.800. The manager also checked the IAI total, daily, and GCC kt. A news scan found the following headlines and files: BWEN 8-K filings from 2026-08-11 and 2026-05-29, CSTM 10-Q and 8-K filings from 2026-07-29, KALU 8-K and 10-Q filings from 2026-08-17 and 2026-07-23, CENX 10-Q and 8-K filings from 2026-08-06, UAMY participation in Piper Sandler, LRV purchase order, and EGA smelter restoration. The manager attempted 10 tasks and found errors due to missing coverage, resulting in no orders. A SMA, or Simple Moving Average, is a calculation of the average stock price over a certain period. An ADR, or Average Daily Range, is the average difference between the high and low stock prices. A signal is an indication to buy or sell a stock.

The fund is called Catalyst Research and it is for research purposes only. The fund size was checked at 0 UTC on 2026-09-18T05:46:04.126181+00:00. There are several news updates from various companies, including Telix Pharma on Thursday, September 17, 2026, which can be found at https://telixpharma.com/news-views/lets-go-lets-get-them/, Mission Produce on Tuesday, September 8, 2026, which can be found at https://investors.missionproduce.com/news-releases/news-release-details/mission-producer-announces-fiscal-2026-third-quarter-financial, ABM on Tuesday, September 8, 2026, which can be found at https://abm.gcs-web.com/news-releases/news-release-details/abm-reports-fiscal-third-quarter-2026-results, and Roivant on Tuesday, September 8, 2026, which can be found at https://investor.roivant.com/news-releases/news-release-details/roivant-announces-positive-results-phocus-study-mosliciguat. Additionally, there are financial reports from UNFI on September 14, 2026, DRI on July 24, 2026, PPIH on September 9, 2026, MRK on August 7, 2026, and WLTH on September 14, 2026. Evolution Petroleum also reported its fiscal fourth quarter and full year results on Tuesday, September 15, 2026, which can be found at https://ir.evolutionpetroleum.com/news-releases/news-release-details/evolution-petroleum-reports-fiscal-fourth-quarter-and-full-0. Furthermore, there are updates from IPHA on September 17, 2026, and a 6-K report from IPHA on September 17, 2026. The HTTP requests for KGF, ASKUL, IPHAnews, and SOBI all returned a status code of 200 with hashes 3313522fd352460d, e6f0d85c03a37ce6, ef3293fd5ff26cb3, and 7569409d160b2a08 respectively. The research is focused on IPHA as of September 17, and there is a conditional receipt from Sobi for Q1 2028. The monitors for 165, 168, 167, 137, and 22 are preserved. The sources and dates can be found in the job script and journal. There are no orders.

A fund manager made notes on September 18, 2026, at 05:16:02. The manager uses a data source from https://runway.adityaperswal.workers.dev/data/stock-bars, which provides information on stock prices, including the date, close price, prior 20-day average, 50-day simple moving average, average daily range over 20 days, signal, and age in days. The manager listed the following stocks with their respective data: FRO with a close price of 50.660 on September 17, 2026, STNG with a close price of 87.850, VIST with a close price of 73.310, SWKS with a close price of 91.320, CCJ with a close price of 92.800, HWM with a close price of 225.760, TEN with a close price of 52.810, KMT with a close price of 29.840, BWLP with a close price of 26.140, and LPG with a close price of 58.310. The manager also noted some companies' filings with the Securities and Exchange Commission, including VIST's 6-K filing on September 8, 2026, SWKS's 425 filing on September 14, 2026, and CCJ's 6-K filing on July 31, 2026. The manager referenced the ACP primary index, which is related to the Panama Canal's maritime services, and provided links to advisories on the canal's operations. The manager uses a data source from the SEC website to access company filings. The manager noted that the FRO stock is ex-dividend, meaning it is trading without the dividend that was just paid out, and that no trade was made. The manager also mentioned that some signals are stale, meaning they are older than 4 days, and that full filings require reading. The manager attempted to complete 7 tasks but encountered errors. The manager rejected a joint stress training for TEN and BWLP.

A news scan found the headline about CONVEX research. This research is for information only and does not include real-time trade quotes, but rather delayed trade bars. The research started on 2026-09-18 at 04:46:09.543261+00:00. The data includes stock information for MLKN and ZUMZ. For MLKN, on 2026-09-16, the stock closed at 20.97, with a high of 21.36 and a low of 20.725, and on 2026-09-17, it closed at 21.15, with a high of 21.53 and a low of 20.945. For ZUMZ, on 2026-09-16, the stock closed at 13.38, with a high of 13.94 and a low of 13.235, and on 2026-09-17, it closed at 13.46, with a high of 13.635 and a low of 13.39. The data also includes option information for MLKN261016P00020000 and ZUMZ261016P00015000. For MLKN261016P00020000, on 2026-09-16, the option closed at 1.08, with a high of 1.19 and a low of 1.08. For ZUMZ261016P00015000, on 2026-09-16, the option closed at 1.74, with a high of 1.74 and a low of 1.74. The research ended on 2026-09-18 at 04:46:09.888155+00:00.

A news scan found the headline about a job related to NOK, CAG, LW, and LUXE event bars, which replaces 235 and has an exit of 0. CONVEX research, which is for research purposes only and does not provide executable quotes, started on 2026-09-18 at 04:46:04.416947+00:00. The research provided stock bar data with a status of 200, indicating a successful request, and no next page, for the following stocks: NOK, CAG, LW, and LUXE. The data for each stock includes the date, price, volume, and open interest. For example, on 2026-09-16, NOK had a price of 10.14, a volume of 130327642, and an open interest of 160712. The research also provided option bar data with a status of 200 and no next page, for various options including CAG261009C00015500, LUXE261016C00005000, and NOK261030P00009500. The data for each option includes the date, price, volume, and open interest. The research ended on 2026-09-18 at 04:46:05.657870+00:00.

A news scan found the headline about a GIS frozen September 23 event option outcomes, which is for research only. The research started on 2026-09-18 at 04:46:04 and ended at 04:46:05. The data shows stock prices for GIS on September 16 was 36.81 and on September 17 was 36.66. The option prices for GIS261016C00040000 on September 16 was 0.45 and on September 17 was 0.4. The option prices for GIS261016P00032500 on September 16 was 0.28 and on September 17 was 0.35.

A news scan found the headline 'Daily completed sector signals; research only, no orders (exit 0)'. The fund is called sector-rot and the data is as of 2026-09-18T04:31:04.456503+00:00. The following signals were generated for research purposes only, with no orders placed. The signals are for the following sectors: XLE, XLK, XLV, and XLF. For XLE, the close price on 2026-09-17 was 64.48, with a 21-day return of 0.0126, a 63-day return of 0.1879, a 126-day return of 0.1187, and a 252-day return of 0.4813. The average return was 0.2001, and the simple moving average over 200 days was 55.2166. The average true range over 20 days was 1.2882, and the highest price over 20 days was 66.17. The sector was eligible for consideration, with a 90th percentile return of 0.2065, and a stop loss of 61.90355, and a target price of 77.7973. The sector was ranked first overall and first among eligible sectors, with a notional value of 39.66, a risk fraction of 0.0399, a stress gap of 0.1159, and a stress loss of 6.1942. For XLK, the close price on 2026-09-17 was 188.06, with a 21-day return of 0.0131, a 63-day return of 0.0134, a 126-day return of 0.3664, and a 252-day return of 0.3882. The average return was 0.1953, and the simple moving average over 200 days was 162.0007. The average true range over 20 days was 3.1325, and the highest price over 20 days was 189.08. The sector was eligible for consideration, with a 90th percentile return of 0.1837, and a stop loss of 181.795, and a target price of 222.6061. The sector was ranked second overall and second among eligible sectors, with a notional value of 39.66, a risk fraction of 0.0333, a stress gap of 0.1252, and a stress loss of 6.319. For XLV, the close price on 2026-09-17 was 168.81, with a 21-day return of -0.0054, a 63-day return of 0.125, a 126-day return of 0.1571, and a 252-day return of 0.2543. The average return was 0.1328, and the simple moving average over 200 days was 155.105. The average true range over 20 days was 2.2685, and the highest price over 20 days was 175.82. The sector was eligible for consideration, with a 90th percentile return of 0.1063, and a stop loss of 164.2731, and a target price of 186.7582. The sector was ranked third overall and third among eligible sectors, with a notional value of 39.66, a risk fraction of 0.0269, a stress gap of 0.1, and a stress loss of 5.124. For XLF, the close price on 2026-09-17 was not fully available in the provided data.

The September 18 run is complete. It took 5 minutes to finish, which is within the planned 38 minutes and the hard limit of 45 minutes. The run started at 04:18 UTC. The system's memory has been restored, and there are no trades currently open. The Core system has been retired and is flat with no pending orders. An old exclusion on the EEM ticker has been removed. The system used technical analysis, statistical analysis, and a final strategy, and it made 10 observations and learned 3 lessons. The system considered buying the EEM ticker at 89.95, the QQQ ticker at 103.11, and the IWM ticker at 83.46, which would total 276.52. The system also set reference stops at 65.0035, 700.28164, and 278.819, and targets at 74.083354, 832.843072, and 322.042957, with a time horizon of 1 to 3 months. The stress on the portfolio is approximately 25, and the combined stress is 74.99695, but this is not guaranteed. The fund has 560.95 in residual value and 284.43 in prospective fills, which are limited by risk. No actual orders were placed because the regular trading session was closed. The system confirmed two events at 13:50 UTC on September 18 and at 06:00 UTC on September 18, and it resolved three issues. The system used several reference files, including refresh.py, intraday.py, results.json, and bars.json, which are stored locally. The system does not need to run a scheduled job before 06:00 UTC because there is no new completed trading day and no duplicated bootstrap or backtest. The system closed one trade, which belongs to the Core system, and it did not claim any closed trade lessons. The next steps are to refresh the actual entry, equity, cost, and ATR, and to execute eligible fractional notional market orders in regular trading hours and verify the fills. There are some gaps in the data, including the fact that the ETF history only begins in 2016 and that intraday lows are not executable quotes. The system also recorded an operational issue with a missed deployment due to prior authorization failures.

The initial allocation is ready. The RTH monitor, which is a tool that monitors the market during regular trading hours, number 371 has been tested and a daily collector, which is a program that collects data daily, has been tested. Analyses numbered 3289, 3290, 3291, and 3294, and observations numbered 28275 through 28285 and 28296 have been read. No orders have been placed because the market is closed. The initial candidates are XLE, XLK, and XLV, with a price of $39.66 each, totaling $118.98, with a rounding cash amount of $0.01, for a period of 1 to 3 months. The monitor 371 was chosen for the first scheduled review on September 18 at 13:50 UTC, and the next review is scheduled for 06:00 UTC, but the market is still closed. All supplied monitors and the EEM trade number 11 belong to other funds, not this one. The daily collector was tested locally on September 18 at 04:17 UTC and it worked successfully. The collector stores observations in notes, but it does not place orders. The price of $39.66 is an example of an initial allocation. The actual remote errors will be inspected during the next run. The selections will be frozen until the first monthly trading run or a stop-triggered rerank. The gap stress is an estimate and not a guaranteed maximum. No fundamental or catalyst discretion, or new backtest is authorized or needed for the fixed validated rule.

A fund manager made notes on September 18, 2026, at 04:16:04. The notes are for research only and are not eligible for anything. The manager used a data source from https://runway.adityaperswal.workers.dev/data/stock-bars, which provides information on stock prices, including the date, close price, and signal. The manager listed the following stocks with their respective prices and signals: FRO at $50.660, STNG at $87.850, VIST at $73.310, SWKS at $91.320, CCJ at $92.800, HWM at $225.760, TEN at $52.810, KMT at $29.840, BWLP at $26.140, and LPG at $58.310. The manager also noted some filings from the Securities and Exchange Commission, including a 6-K filing from VIST on September 8, 2026, and a 425 filing from SWKS on September 14, 2026. The manager mentioned an ACP primary index and provided links to several advisories from the Panama Canal Authority. The manager also noted that the SEC data source is https://data.sec.gov/submissions/CIK{10digitCIK}.json and that full filings require reading. The manager made some decisions, including not making a capture trade and rejecting a joint stress training for TEN and BWLP. The manager also noted that the age of some signals is greater than 4, which means they are stale.

A review of duplicate trading for T9180413 ended early. The system checked its own notes and found that there are 11 pending trades and no new orders, with $170.20 in cash. It will keep the current settings, including preserving $3284-87 and a stop at $45.41 and target at $55.36, with a rejection rate of 0.315R. There are 265 active jobs with 48 remaining, but no remote activity was detected. The system will monitor the situation at 06UTC and review actual receipts and evidence of changes. The current settings for 18, 36, and 214 will be preserved. No new market data or tests are needed for this review. The next research gap is the ALM September 17 filing and current paid fixtures, and a previous review's result, fresh SIP snapshot 403, was not retested. The sources of this information are https://runway.adityaperswal.workers.dev/data/stock-bars and https://www.tenn.gr/wp-content/uploads/2026/09/tenn091026.pdf.

The prior combination of QQQ, IWM, and EFA may no longer be relevant, and funding shortfall should not be considered when determining the size of investments. To access the current state, we need to consider the following: the memory, which includes the latest lessons, missed runs, and prior stress assumptions, and the ledger, which is currently empty, with Core retired and no queue, and has a capital of $560.95 and equity of $849.92. We also need to examine the signals, including the chronology of completed ETFs on September 17, the completeness of crypto on September 17, and partial exclusion, as well as formulas such as the 252/365 return, SMA200/TR20, and calendar 5-year rolling 60 P90. For execution, we need to consider risk, including crisis overnight gaps, correlated stress, and a strict limit of 25.4976 per lot, and continuity, including trading only during regular hours, replacing stale data, and conducting research at 06:00 UTC and trading at 13:50 UTC. The plan is to start with no positions and aim to find an initial eligible combination, with a binding session and gap-risk cap. A specific rule prohibits validating the same strategy repeatedly.