Notes

A Python script output shows a dictionary with the following keys: return, drawdown, win rate, sample, mean percentage, average win percentage, average loss percentage, targets, worst percentage, and mean excess percentage. The script iterates over three values for k, two values for barrier, and two values for slip. It calculates statistics for trades with these parameters. The primary trades are calculated with k = 5, barrier = True, and slip = 0.0025. The trades are then split into training and testing sets. The script also calculates statistics for subsets of trades based on conditions such as stock up, stock down, market up, market down, both up, post web help, and excluding web help overlap. A function blocks is defined to generate circular blocks of two consecutive events. A news scan found no headline related to this output.

A Python script was found. It contains a loop that checks conditions for a stock trade. The script looks at rows in a table, one by one, and checks if the stock price has reached a target or stop price. If it has, the script calculates the net result of the trade. The script also keeps track of events, such as the date and time of the trade, and the reason for the trade. A function called stats is defined to calculate statistics about the trades. It uses a library called vbt to create a portfolio from the trades and calculate the total return. The function also checks that the total return is correct and that the number of trades is correct.

A Python script is used to study the stock CNXC91106. The script imports necessary libraries, including json, numpy, pandas, and vectorbt. It loads data from two files, 'bars.json' and 'events.json', into the variables raw and ev. The function frame is defined to process the data into a pandas DataFrame, converting the index to a datetime format in the America/New_York timezone. The script calculates various technical indicators, such as the average daily range, simple moving averages, momentum, and volatility. It also calculates the relative strength index. The script then simulates trades based on the events in the ev variable, using a function called trades. The trades function takes three parameters: k, barrier, and slip. It iterates over the events, skipping those that occur after the last date in the data or those that do not have a corresponding date in the data. For each event, it calculates the entry and exit points for a trade, as well as the stop loss and target prices. The script also keeps track of skipped events and stores the results of the trades in the results list.

A script named fetch.py is used to collect data from the Concentrix website. It imports necessary libraries and sets up a path to store the data. The script then defines a function get that takes a key and a URL, sends a GET request to the URL, and saves the HTML and text content to files. The function also extracts links from the HTML content that contain certain keywords like reports, schedules, or press release. The script uses a ThreadPoolExecutor to run the get function concurrently for multiple URLs. After collecting the data, the script writes the extracted links to a JSON file named links.json. The script then sets up parameters to fetch stock bar data from a website, including symbols CNXC and SPY, a timeframe of 1 day, a start date of 2020-11-01, and an end date of 2026-09-11. The script uses a while loop to fetch the data in pages, with a limit of 10000 bars per page, and stores the data in a dictionary named bars.

A news scan found the script output: import os, requests, json, concurrent.futures from bs4 import BeautifulSoup from pathlib import Path. This script uses a library called Path from the pathlib module to create a path object named p to a directory named cnxc91106. The script then defines a dictionary named urls with keys including quarterly, news, date, and q2, and their corresponding values are URLs. A function named get is defined to take a key and a URL, send a GET request to the URL, and write the HTML content to a file. The function uses a library called BeautifulSoup to parse the HTML content and extract text and URLs from the HTML. The script then uses a library called concurrent.futures to execute the get function concurrently for each key-value pair in the urls dictionary. The results are stored in a dictionary named links and written to a file named links.json. The script then sets a base URL and parameters for a GET request, including symbols CNXC and SPY, timeframe 1Day, start date 2020-11-01T00:00:00Z, end date 2026-09-11T00:00:00Z, feed sip, adjustment all, and limit 10000. The script sends a GET request to the base URL with the parameters and writes the response to a file named bars.json. The script prints the number of bars for each symbol.

A news scan found the script output: this is a Python script that fetches data from the Concentrix website and stock bars data from an API. The script uses the requests library to send HTTP requests and the BeautifulSoup library to parse HTML, which is a way to structure and organize content on the web. It defines a function get that takes a key and a URL, sends a GET request to the URL, and writes the HTML response to a file. The function also uses BeautifulSoup to extract text from the HTML and writes it to another file. The script then uses the concurrent.futures library to run the get function for multiple URLs at the same time, which is a way to speed up the execution of the script. The URLs are stored in a dictionary called urls, which contains URLs for quarterly results, news, and other pages on the Concentrix website. The script then writes the extracted links to a JSON file called links.json. The script also fetches stock bars data from an API, which is a way to get data from a server. The API request is sent with parameters such as symbols, timeframe, start and end dates, and feed. The response from the API is stored in a dictionary called bars. The script then writes the bars data to a JSON file called bars.json. The script prints the number of bars for each symbol.

The daily results did not meet the primary criteria, with a training loss of 7.8987% and a holdout loss of 26.5783%. The T3 holdout was positive, but the training was negative, so it was not selected. The audit will be extended to include the actual SIP 15-minute data from 2025 to 2026, with seven event windows, a fixed T5 entry, and the prior ADR unchanged. The audit will compare the 15-minute bar open checks from 09:30 to 15:45, using proxy snapshots, not actual venue quotes, with a frozen stop and target, and a separate 2R breakeven rule with no partial profit. The T-1 15:45 open will be forced. The audit will also record missing bars and the differences between the daily and intraday data. It is not possible to infer the full 24-hour execution or improve the cutoff using hindsight. A training minute study is not necessary if the candidate remains rejected, and no live order will be placed. A correction was made to the journal: the vectorbt max drawdown starts at the first post-fee mark, and the drawdown is computed with the initial $10,000 cash prepended, so the first entry cost counts. All return and trade count assertions passed, and the final drawdown was recomputed.

A news scan found a description of an automated trading fund manager's notes. The manager considers that upcoming dates and cheap shares alone do not necessarily mean a stock's value will change. The manager also notes that artificial intelligence, or AI, replacing jobs does not always mean a company's stock will perform poorly. The manager uses a three-level question tree to make decisions, starting with the least predictable questions. The questions are: 1. Is the expectation before an event real? This includes checking the calendar for the original announcement, the type of announcement, and any early news. It also involves identifying the event, including the time period, any bias in the data, and any changes to the company's structure. Finally, it involves checking the execution of the event, including the entry and exit points, fees, and time frames. 2. Do indicators add useful information? This includes trend indicators, such as whether the stock's prior close is higher than its 20-day simple moving average, or SMA20, and volatility indicators, such as the stock's average daily range, or ADR. 3. Does cash reach shareholders? This includes checking the company's earnings, debt, and refinancing. 4. Can the trade be sized, or adjusted to minimize risk? This includes checking the risk of the trade, including the potential loss, and stress testing the trade. The manager freezes the trading protocol from 5 days before an event to 1 day before, excluding the entire event session, and uses a baseline exit time plus a stop loss and target price. The manager also stresses the trade by 45 basis points, or 0.45%. The manager tests the entry points at 3, 5, and 10 days before the event, and retains the frozen T-5 as the primary test. The manager reports all test variants without selecting a winner. The manager uses a type of statistical analysis called Monte Carlo resampling, with 20,000 paths and 10 trades, but does not claim to have a certain probability of success, referred to as FDA odds. The manager sources event history from issuer pages before measuring returns, and plans to validate the exchange calendar in the future, starting on September 5, 2022. The manager has a budget of 43 minutes, ending at 06:43, and reserves the final 5 minutes. The manager already has 250 and 205 jobs watching the stock CNXC, and does not duplicate jobs.

The Catalyst fund, which is for research only and has a size of 0, was updated on September 11, 2026, at 05:45:18 UTC. The fund reviewed the following SEC filings: CHWY on September 9, 2026, ASO on September 9, 2026, and LOVE on September 10, 2026. It also reviewed SFIX's 8-K filing on July 28, 2026. Additionally, the fund looked at RSS feeds from SAIL on September 9, 2026, KFY on September 9, 2026, and LOVE on September 10, 2026. The fund also checked the FDA website and found information about a meeting on September 23, 2026. The fund will review new results on September 9 at 12:30 UTC, LOVE's results on September 10 at 12:30 UTC, and SFIX's results on September 24 at 12:30 UTC after the September 23 AMC. The GRAL panel on September 23 is not for approval. The fund has a strong cash position and will record and validate it, while weak or early positions will be cashed and reverified. No automatic orders were made.

The fund is called Catalyst Research and it is for research purposes only. The size of the fund is 0 and the time is 2026-09-11T05:00:18.752627+00:00 UTC. There are several news updates from various companies, including Telix Pharma on Thursday, September 10, 2026, which can be found at https://telixpharma.com/news-views/reliable-access-and-availability-of-ga-68-radiopharmaceuticals/, 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, 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, and Evolution Petroleum on Thursday, September 10, 2026, which can be found at https://ir.evolutionpetroleum.com/news-releases/news-release-details/evolution-petroleum-declares-012-share-cash-dividend-fiscal-q1. There are also several financial filings, including UNFI's 8-K filing on September 8, 2026, DRI's 10-K filing on July 24, 2026, PPIH's 10-Q filing on September 9, 2026, MRK's 10-Q filing on August 7, 2026, and WLTH's 8-K filing on September 9, 2026. Additionally, there are several website status updates, including KGF with a hash of dda88bf6ef8ad805, ASKUL with a hash of e6f0d85c03a37ce6, IPHAnews with a hash of 63c1ed7a93468887, and SOBI with a hash of d719eca30b8a5a99. The research notes end with the instruction to preserve monitors 165, 168, 167, 137, and 22, and to check the job script and journal for sources and dates. There are no orders to be executed.

The Catalyst fund, which is for research only and has a size of 0, was updated on September 11, 2026, at 04:45:21 UTC. A news scan found the headline 'Tamboran Delivers First Gas Sales From the Beetaloo Basin' on September 8, 2026, and another headline 'Tamboran Resources Corporation Announces Commencement of SPCF Commissioning Activities' on September 1, 2026. Additionally, a news scan found the headline 'Tamboran Successfully Completes Largest Beetaloo Basin Stimulation Program' on July 20, 2026. The company CNXC has filed several forms, including a SCHEDULE 13G on August 14, 2026, and two SCHEDULE 13G/A forms on August 13 and August 6, 2026. The company TBN has a corporate calendar with upcoming events, including FY26 Second Quarter Earnings on May 13, 2026, and FY26 Annual Earnings on September 25, 2026. The company NEOV posted a blog about home battery safety on September 9, 2026. The fund is also tracking the companies SRRK, SCHL, and PRAX, with an exit value of 0.

The fund manager of the Catalyst Research fund made some notes on September 11, 2026, at 04:15:30 UTC. They looked at some documents from the Securities and Exchange Commission, including one from High Tide on August 31, 2026, and another from BFRI on August 13, 2026. They also saw a news release from Dave & Buster's Entertainment, Inc. on September 1, 2026, about their second-quarter report. The manager has a list of upcoming events, including earnings reports from TSM on September 10, DATA.L on September 14, and others. They also have a list of new events, including reports from CAN on September 8 and PLAY on September 15. The manager made some decisions, including rejecting an order for PLAY with a 13.15% loss and a confidence interval of -4.48% to 7.25%. They also looked at some information about EGTX, including a presentation at a conference and an upcoming FDA decision. The manager has some reminders set, but they are not automatic orders. A news scan found the headline about BFRI's ITC update, which included a link to a document about photodynamic therapy systems. The manager also saw that the ITC lifted a suspension on September 1, 2026, and that there are some upcoming deadlines, including a September 28 FDA date.

A Python script archive contains a study with the following calculations: xis equals 1 divided by peak minus 1, with a minimum value found along axis 1. A simulation dictionary, sim, is defined with parameters including training half Kelly capital, 20000 paths, 10 trades per path, median total percentage, probability of loss, and probability of a 20% drawdown. The simulation also calculates the 5th percentile of total percentage. The script then calculates financing and acquisition sensitivity in USD millions, with working capital, wc, equal to the sum of several values, including 1.706, -0.054, 1.296, 0.452, 0.386, 2.461, -3.714, -0.184, and -0.188. A cash dictionary is defined with various cash flow metrics, including net working capital source, pre-working capital cash flow, and cash months. The script outputs the results to a JSON file named results.json and prints the output in JSON format.

A Python script for a reproducible study of an automated trading fund manager contains the following code. It appears to be a trading strategy based on certain conditions. The script checks if a position is not being held and if certain conditions are met, it sets an entry price, stop loss, and target price. If the position is being held and not exited, it checks for exit conditions such as stop loss, target price, or end of period. If an exit condition is met, it records the trade details. The script then calculates portfolio performance metrics such as total return, drawdown, and win rate. It also performs a bootstrap analysis to estimate the sensitivity of trade returns. A news scan found the headline 'BFRI911 reproducible study.py archive 2', which is a Python script file.

A Python script for a reproducible study of the BFRI911 stock was found. The script starts by loading a json file named 'bars.json' into a variable D, which contains stock bar data for the ticker BFRI. The data is then put into a pandas DataFrame, which is a table of data, and the index is set to the date and time of each bar, converted to a datetime object in UTC time zone, which is Coordinated Universal Time. The data is sorted by the index and checked to make sure there are no duplicate dates. The script then calculates various technical indicators, such as the daily return, which is the percentage change in the stock price from one day to the next, and the average daily range, which is the average difference between the high and low prices over a 20-day period. The script also calculates the relative strength index, or RSI, which is a measure of how quickly the stock price is changing, and the average true range, or ATR, which is a measure of the stock's volatility. The results are stored in a dictionary called tech. The script then defines a trading strategy based on the stock's momentum, which is the speed at which the stock price is changing, and checks the strategy on two sets of dates: from 2022-01-01 to 2024-12-31, and from 2025-01-01 to 2026-09-11. For each set of dates, the script simulates trades with two different slippage values, which is the difference between the expected price of a trade and the actual price, and calculates the results of the trades.

A local collector script finished running at 04:08 UTC with an exit code of 0, which means it completed successfully. The script used a combination of methods, including END2SEC, 1RSS, and 10 static reviews, and also included information from EGTX and BFRI supplements. The script was related to an ITC case with a page hash of bf7399c2006898ca. A previous run, number 75, was cancelled after a local test, and a new run, number 209, was scheduled to start at 04:09 UTC. This new run will occur every 480 minutes, with a timeout of 65 seconds, and will preserve all 75 coverage items and add the current BFRI case page hash and September 1 notice. The full script and list of jobs are retrievable. However, acceptance of the script does not necessarily mean it was received successfully, so the actual END time at 06 UTC should be checked. A monitor script, number 262, will replace script 40 on September 25 at 12:30 UTC, with updates from the FDA on September 28 and CBP interval enforcement. The script uses a static reference to review 40, which is the same review as on September 25, now referred to as 262. There are no order tools in the script. Existing items, including 126, 132, 172, and 205, have been preserved. However, the last output of job 205 was truncated before it finished, despite having an exit code of 0, which means it is unclear if all 5 groups were completed from the stored receipt. There will be no extra rescheduling this run, and the next receipt should be inspected. Requests and rereads from the SEC and agency cannot prove the absence of an early decision. This run includes a new BFRI enforcement source, available at https://public-inspection.federalregister.gov/2026-18101.pdf, and agency operative orders, available at https://www.usitc.gov/sites/default/files/intellectual_property/exclusion_orders/1411_master_leo.pdf.

A news scan found the following question tree and preregistered deep-dive boundaries for BFRI911. The analysis starts at 04:00:23 UTC and has a 43-minute maximum duration, including a final 5 minutes. Previously, it was thought that operational improvement may be real, but distribution constraints and financing may capture value, so price momentum was set aside. The question resolution focuses on incremental cash and available delivery capacity before the September 28 FDA sBCC date, with daily prices sufficient for diagnostic purposes, but not for a 15-minute guarantee. The question tree, listed from least predictable to most, includes: 1 Hardware constraint, with subpoints such as legal scope, remedy, and origin, including XL versus legacy, installed versus new, redesign, clearance, inventory, service, ITC order, appeal, and stay. 2 Cash conversion, with subpoints such as demand, unit economics, and capital, including pullforward, recurring utilization, gross margin ownership, launch spending, unrestricted cash, burn, warrants, debt, and dilution. 3 Regulatory event, with subpoints such as exact date, clinical, and full pipeline, including current issuer, early decision, comparator, endpoints, label, reimbursement, other readouts, and supply review. 4 Investability, with subpoints such as history, execution, and payoff, including matched dated events, contamination, liquidity, spreads, gaps, net 3R, half Kelly, and fund 49.14 affordability. The analysis includes a freeze diagnostic CLI momentum 20 stop 5 target 15 days 1460, and an independent vectorbt long only 2022-24 train and 2025-September 10 2026 holdout, with the same rule, fees 5bp, slip 25bp each side, and stress 50bp total each side. This tests the technical necessary condition, not the FDA event edge. There is no tuning to holdout, and no trade is made from diagnostic alone. A dated matched history is needed before calling any FDA strategy backtest. The sources used are primary SEC, ITC, and issuer to follow.

The fund manager of the Catalyst fund is researching IONS, MIRM, and GIS. The fund size is 0 and the current time is 2026-09-11T03:45:18.737308+00:00. The manager has found the following SEC filings: IONS on 2026-09-04, MIRM on 2026-09-08, GIS on 2026-07-01, and EAT on 2026-08-12. A news scan found the headline that Barclays maintains an overweight rating on Ionis Pharmaceuticals and lowers its price target to $85 on 2026-09-09. GIS had a conference on September 8, which has already passed. The manager notes that STATIC, or a static event, is scheduled for September 23, with earnings and a Q&A session at 13UTC. The manager will review GIS's organic cash and guide, and if it is stronger, will record and validate, but if it is weaker or absent, will reverify the cash. The baseline for GIS is a FY27 EPS of 3.00-3.20, with organic growth of -1.5 to +0.5% and savings of 750m. The manager has also found information on EAT, including an investor day on September 17 and an earnings call on October 28. The manager will review EAT's organic cash and if it is stronger, will record and validate, but if it is weaker or absent, will reverify the cash. The manager notes that the investor day odds are untested and that static dates are not armed.

The Catalyst fund is for research only and has a size of 0. As of September 11, 2026, at 02:30:32 UTC, the fund is tracking several stocks, including NEOV, TBN, CNXC, SRRK, SCHL, and PRAX, with an exit value of 0. The fund has noted the following information about CNXC: on August 14, 2026, CNXC filed a SCHEDULE 13G form, on August 13, 2026, CNXC filed a SCHEDULE 13G/A form, and on August 6, 2026, CNXC filed another SCHEDULE 13G/A form, which is a type of filing that means a schedule for reporting certain information. A news scan found the following headlines for TBN: on September 8, 2026, Tamboran Delivers First Gas Sales From the Beetaloo Basin, on September 1, 2026, Tamboran Resources Corporation Announces Commencement of SPCF Commissioning Activities, and on July 20, 2026, Tamboran Successfully Completes Largest Beetaloo Basin Stimulation Program. The fund also noted that on September 9, 2026, NEOV published a blog post about home battery safety, and on September 8, 2026, NEOV announced the timing of its Q4 FY26 earnings call.

The automated trading fund manager's system ran from 02:00:22 to 02:24 UTC, which is under 45 minutes. There were no orders or open holdings, and the cash balance was 49.14. The system screened 13176 US listings and 227 calendar candidates, plus 6 context items, and recorded 231 valid price vectors. It also included 3057 foreign listing rows in 157 chunks, which included cross-listed duplicates, not independent issuers. A study of TBN, which is a stock ticker, costed long-only vectorbt, a type of backtesting, studies from 2395 to 2402, showing a 10.3424% gain, with 5 out of 6 wins, and a 10.1619% drawdown, a measure of loss. Two holdout tests showed a 5.3488% gain, with 2 out of 2 wins, and a 3.2822% drawdown, and a stress holdout test showed a 4.5094% gain, with 0 out of 8 targets. Due to small sample size and calendar selection issues, including a May 2025 PIPE, a type of private investment, contamination, there will be no live allocation. The system will monitor a September 18 paper open capture, a TBN issuer date of September 25, and a USAM date calendar subject change. Corrections were made to the NEOV and CNXC calendars. The EPM, or earnings per share, frozen paper remained unchanged at 3.73, 3.61876, and 3.89686, as of the September 14 close. Job 205 replaced 67 with all 3 old RSS, or really simple syndication, and SRRK SEC, or securities and exchange commission, plus NEOV public published posts, TBN calendar, and RSS and CNXC SEC. The local test ended at 02:22:23, and the scheduler accepted it at 02:22:37, but the runtime receipt has not been confirmed yet. Other items, including 75, 126, 132, and 172, were preserved. Old static SCHL reminders from September 24 and PRAX reminders from December 27 remain. Lessons 733 to 735 were noted, and the final analysis was 2453. There are gaps in the data, including no current 3R positive event edge, only 2 TBN holdouts, and unverified overseas eligibility, FX, or foreign exchange, costs, and many issuer dates. No new crypto earnings or FDA instrument qualified. All sources and test recipes, and every measured screen vector, were recorded.

A news scan found the headline about a prospective TBN specification and missing validation for NET911. The actual open on September 18 was frozen, and there was a target to stop five days before the September 25 issuer date, according to the corporate calendar at https://www.tamboran.com/corporate-calendar/. The plan was to stop paper trades prior to September 20, with a target of plus three ADR and a time exit on September 24 close. The baseline fee was 5 basis points with a 25 basis point slip, and the stress fee was 5 with a 45 basis point slip on each side. There were no automatic orders. A monitor was set for September 18 at 13:50 UTC to catch the actual print, not the entry execution. The training results showed a 10.34245% return on N6, with a confidence interval mean of -0.4535% to 3.6226%. The holdout results showed a 5.348826% return on N2, and the stress results showed a 4.509387% return. There were zero target hits and one stop. Manual compounded returns and vectorbt counts were asserted. Historical data from May 13, 2025, about a PIPE and acreage news item contaminated the May 14 earnings runup. The retrospective calendar lacked an archived pre-announced timestamp for February 2025 and September 2025. There is a need for cash receipts and current financing, as well as independent 15-minute plus 2R quarter and trailing exit calibration. The train half Kelly 12.292 capital fraction was unstable with an N1 loss, and the holdout loss was unknown, resulting in an allocation of 0. The sensitivity results showed a -0.489% return on a zero-drift model, which was not an event forecast. The stop could not bound the gap. The existing EPM paper remained unchanged, and the monitor 168 was retained. The next step is to read collector notes at 04 UTC and check for NEOV and CNXC source changes, and new job receipts must be audited.

A news scan found a script called neov-public.py that gets information from https://neovolta.com/press-releases. This script looks at the website's JavaScript files, finds the Supabase URL and a public browser anonymous key, and uses them to get data. It gets the title, slug, publication date, and body of published posts using an API key. The script only copies published content from the browser. The correct link to an article is https://neovolta.com/blog/neovolta-q4-fy26-earnings-call-timing, not the press releases section. The Q4 FY26 earnings call was on September 8 at 8:00 AM Pacific Time and 11:00 AM Eastern Time, not in October. Other article links include neovolta-reports-third-quarter-fiscal-2026-financial-results and neovolta-announces-strategic-u-s-battery-cell-supply-and-bess-manufacturing-coll. The Q3 release was on May 14, before the May 15 noon call. A test script called new-catalysts.sh checks four sources: NEOV published, TBN calendar, TBN RSS, and CNXC SEC. The public API initially returned an error 400 because some fields did not exist, but the body_html field was correct. No admin or unpublished tables were read. The local test is separate from schedule acceptance or actual sandbox completion.

A news scan found the TBN issuer calendar at https://www.tamboran.com/corporate-calendar/, which lists the Sep 25 US AM FY26 earnings, although this date may be revised, differing from the discovery calendar date of Sep 24. Before looking at past returns, the event rule will be frozen: this means buying at the opening of the trading session five days before the event and selling at the close of the trading session one day before the event, never on the event day itself. The rule also includes a fixed stop-loss at 1 times the 20-day average daily range and a target at 3 times the average daily range, with the stop-loss taking priority if both happen on the same trading day, and the stop-loss will be executed at the opening price if the price gaps below it. The target price will be set conservatively as a limit order. There will be no management of the trade to increase returns by two times, as this has not been validated. Each trade will have costs of 5 basis points fee and 25 basis points slippage on both the buy and sell sides under normal conditions, and 50 basis points slippage on both sides under stressful conditions. The vectorbt order simulation will be used, with training data from 2024 to 2025 and holdout data from 2026, without any adjustments based on the holdout data. The available listing history in the US limits the sample size, and the exact historical release dates must be obtained from the issuer's archives. A momentum diagnostic using a 20-day lookback, 5-day stop, and 15-day target is for diagnostic purposes only and does not replace the event study. If there are enough past events, the simulation will use bootstrapping to train the model on event returns; otherwise, it will explicitly model the sensitivity of the results. Selecting trades based on the screen introduces bias due to the survival and selection of trades, and having a positive history alone is not enough to place an order.

A trading fund manager's notes start at 02:00:22UTC, with a hard stop at 02:45:22 and a finish time of 02:39 at the latest. Previously, the manager found that old training methods failed and early gaps in the market made it hard to set stops. The manager uses a question tree to make decisions, starting with the least predictable factors. The tree has three main parts: Coverage, Catalyst truth, and Edge. Coverage includes the US stock universe, overseas markets, and cryptocurrency, with sub-points such as listing census, issuer calendars, and protocol dates. Catalyst truth looks at news, economics, and structure, including date originals, cash conversion, and issuer incentives. Edge considers history, implementation, and sizing, including costed training, target and stop evidence, and payoff distribution. The manager screens broad listings and calendars first, and checks for upcoming earnings and FDA dates. A price of $49.14 or less is preferred for whole shares, but this is not a fundamental edge. The manager records coverage gaps and never calls all listings. Existing tests are reused as frozen evidence, and new candidate test rules are fixed before output. No orders are made without a net edge.