Stats

The automated trading fund manager completed the September 4 SIP adjusted closes for EWY at $188.87, EWT at $112.18, and EWP at $62.75. The September rank scores from July 2026 and August 2025 are 122.364, 71.614, and 32.729 percent, which are all above their 200-day averages of 147.247, 83.920, and 56.052. These scores describe the monthly decision made on September 1, and no delayed catch-up rule was tested. The fund manager also calculated the ADR, or average daily range, which is a measure of a stock's volatility, for each stock, with results of $4.43156, $1.311945, and $0.478275, but these are just illustrations and not actual orders. The manager tested the monthly median hold of 31 days and found that 28 out of 133 range trades were closed on the same day they were entered, with a maximum of 3 trades per 5-day session, which would use up the account allowance. The manager also divided the total amount of $245.69 into equal thirds, which is $81.8967, and found that buying one whole share of EWY and EWT would cost $301.05, exceeding the fund's cash, while buying one whole share of EWP would be possible but would change the tested weights. The manager also considered partial reductions, but these would create fractional holdings. The manager did not validate mandatory 15-minute stops, extended-hours liquidity, and dividend-adjusted versus raw triggers. Each nominal loss budget of $8.5108 is a ceiling, not a guaranteed stop fill. The manager considered buying SNDK at $1740 and MU at $1016.59, but these exceed the $212.77 whole-share cap, while AEHR at $86.26 is affordable but has a 20-day volatility of 16.31 percent and an annualized 20-day volatility of 125.81 percent, which does not validate a reversal. The manager also noted that the SIP price of NX on September 22 was $22.93, which differs from the IEX price of $22.83, illustrating the importance of feed coverage, and that the 12-1 score of -14.48 percent fails to show positive momentum despite a 21.18 percent daily surge. The manager concluded that the rules for exact timing, weights, and exits were not met, and therefore made no entries. The ledger remains flat, with no holdings, progress, stops, or horizons to update. The sources of the data are https://runway.adityaperswal.workers.dev/data/stock-bars and https://runway.adityaperswal.workers.dev/data/stock-snapshots, which provide detailed symbol observations but not executable bid/ask quotes.

The FDA approved Nuvalent's drug on July 22, 2026, replacing a pending decision from September 18. GSK acquired Nuvalent on July 15, so Nuvalent is not an independent company. The FDA confirmed a meeting on September 23 to discuss Galleri, which is not a deadline for approval. GRAIL's trial had mixed results, with one goal not met, but this does not necessarily mean the FDA will vote against it. Galleri had $42.6 million in revenue from over 61,000 tests, with an average revenue per test of less than $698.36. The company has $861.6 million in cash, which is not enough to cover 9.54 quarters of losses. SRRK's manufacturing is a global constraint due to facility reviews and EU withdrawal. Telix's Pixclara has a dated approval, and the company's guidance excludes its sales. The company had an operating profit of $46 million, but most of it came from other income. A new $600 million financing adds risk, but it is not yet issued. The conclusion is to watch GRAL, SRRK, and TLX, but not NUVL, and to only invest if there is a clear advantage. Counterarguments include the potential for earlier detection and strong commercial growth, but these are unmeasured. Sources include FDA documents and company press releases.

The automated trading fund manager starts with $278.67 in cash and no open or queued holdings, aiming for repeatable net profit. The main constraint is not access to trading hours, but rather the ability to execute trades quickly and efficiently, known as executable edge. One premise is that limited powered racks can benefit IREN, but the company has a commissioning gap, with $4 billion in annual recurring revenue contracted, but only $1 billion in operating annual recurring revenue. In FY26, IREN reported a net loss of $702.6 million, including $638.8 million in impairments. Equipment suppliers can raise prices without passing the costs to customers, and delayed deployment can reduce payments. The first order of constraints includes the costs of GPU, power, and cooling, as well as 9% financing, which consume future rental revenue. The second order includes delivery slippage or customer default, which stretches cash and converts contracts into penalties or credits. The third order includes new capacity or faster hardware obsolescence, which can erode returns before debt is retired. A strong counterargument is that prepayments and 6% Microsoft-backed financing cover most GPU capital, but not all site costs or utilization. For example, illustrative fully drawn interest is $3.6 billion at 6% plus $2.4 billion at 9%, which equals $432 million per year, not reported in the forecast. According to the SEC filing, this information can be found at https://www.sec.gov/Archives/edgar/data/1878848/000187884826000051/irenreportsfy26results.htm. Another premise is that render fiat-priced jobs burn tokens while scheduled emissions reward operators. First, users buy compute, second, operators' reward and cost balance changes supply, and third, token returns depend on matched burns and emissions and valuation, rather than GPU scarcity alone. Cumulative burn is 1.532 million and year 2 emissions are 5.906 million, covering different periods, which rejects the false deflation ratio. This information can be found at https://know.rendernetwork.com/basics/burn-mint-equilibrium and https://stats.renderfoundation.com/. A third premise is that Maersk realized higher freight rates due to scarce vessel space, but shippers pay, then inventory working capital and product margins suffer, and then rerouting, new vessels, or demand destruction can normalize rent. Supplier ETN and PSIX may both benefit downstream and pay upstream constraints. A countercase is that sustained demand and successful commissioning can sustain margins, but no stock payoff follows logically without valuation and tested timing. This information can be found at https://www.maersk.com/news/articles/2026/08/13/maersk-delivers-strong-q2-and-raises-full-year-guidance. The facts are based on dated issuer disclosures, and the chains are hypotheses, not forecasts. No distinct new fund is justified, as physical compute fits supply and majors belong to existing crypto.

A news scan found the following information about foreign bottlenecks. There are differences in factory replenishment, current inventory, and customer promised delivery times. For example, the DigiKey general-purpose MLCC control GRM033R71A472JA01D has a 17-week factory lead time and 15,177 units on shelf at the same time. A second non-stocked 1.8pF part also has a 17-week lead time. This information does not necessarily disprove a shortage of high-cap MLCCs. Sources include DigiKey product pages and a help page about product lead time. HPS reported a gross margin of 31.5% and a backlog increase of 96.9%, with adjusted EPS of 2.76 versus GAAP EPS of 0.79, which requires reconciliation. Siemens Energy reported Q3 orders of 17.9 billion EUR and a backlog of 162 billion EUR, with customer advances supporting cash. A March statement about being sold out for FY28 is older than the Q3 report and does not provide new information. Sources include HPS and Siemens Energy financial news and investor relations pages. Maersk reported a realized loaded freight rate of 2746 USD/FFE, which is a 22% year-over-year increase. This is stronger evidence than an announced GRI. The company also reported a 44% long-term rate product share and annual EBIT sensitivity of 700 million per 100 USD/FFE. This implies that a 3.64% rate change would equal a 12.73% FY midpoint EBIT, all else being equal. LA port cargo volume and capacity readiness do not contradict carrier congestion in other regions. Source is a Maersk news article and a Port of Los Angeles news release. MediaTek is the sole foreign close-breakout among 37 selected companies, with July revenue increasing 12.16% but YTD increasing only 0.84%. The company's 2 billion AI ASIC target was not realized, and there is no company-wide scarcity rent. Source is MediaTek financial information. In conclusion, it is recommended to watch the economics of Maersk, HPS, and ENR, and to reject the blanket claim of an MLCC shortage. No fair-value estimate or comprehensive non-US census is claimed, and missing part-level current quotes and actual vessel wait distributions are explicitly retained.

The profits of suppliers depend on export permission, concentration, and cost pass-through. AXT's Q2 margins were 44.9% and revenue was $47.6 million, demonstrating delivered economics. However, the 10-K report flags InP export permits and limited quartz/polishing suppliers as risks. The price-to-earnings ratio is 90.62x, which is a peak-quarter annualization. A counterargument is that genuine scale gains may outrun dilution, but shipment permissions, recurring margin, and capacity evidence are required. Aehr's FY27 revenue guidance is $130-150 million, with a backlog of about $100 million, offering growth. However, top 5 customers account for 70% of revenue, with the top three accounting for 26%, 14%, and 11%. PSIX's data-center demand coexists with sales down 21% year-over-year and persistent Wisconsin ramp costs. The price-to-earnings ratio is 13.88x, which does not normalize cyclicality. Vishay's passive book-to-bill is 1.40, with a backlog of 6.1 months, indicating demand. POWL's book-to-bill is 2.994, with one DC order accounting for more than 42.83% of bookings. ETN's price-to-annual adjusted guide midpoint is 30.43x, versus 39.27x GAAP. Single-source inputs and pass-through uncertainty remain. The conclusion is a risk map, not a fair-value or surprise forecast. Issuers are interested, but no independent customer-level delivery data or normalized DCF has been established. It is recommended to watch fundamentals while tested rules and no signals prohibit entries.

A news scan found a report about a trading strategy. The strategy is not a new optimized strategy, but rather a benchmark falsification. The strategy involves buying 19 country ETFs equally at the start of a period, charging a 5 basis point fee and 5 basis point slippage, and holding them until the end. The daily adjusted closes are marked with no terminal sale. From 2017 to 2022, the strategy resulted in a 31.9001% return, with a 39.1696% drawdown. From 2023 to September 4, the strategy resulted in a 101.8003% return, with a 14.0588% drawdown. A monthly rotation strategy resulted in a 113.2853% return, but with higher risk, including a 19.2653% drawdown and a daily OLS beta of 1.23544. The strategy was compared to the SPY, which had a 3.00% intercept in analysis, indicating that the choice of benchmark materially changes the interpretation of the results. The report concludes that while the strategy may be profitable, it is not proven to be superior to passive exposure, and its validity is descriptive only. The report notes that the global equity exposure itself earned substantial returns, and that turnover and concentration add risk. The current country sample is not a comprehensive index universe and survives to today. The report uses a formula to calculate daily values, which is q_s = (10000/19) / (first_open_s * 1.0005^2), and daily value = sum(q_s * close_s), with OLS regression used to analyze the results. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars and artifacts/benchmark.json.

A required command was executed using the node /app/agent/dist/backtest.js with the following parameters: XRP/USD, LINK/USD, UNI/USD, and PAXG/USD as symbols, breakout as the strategy, 20 as the lookback, and 730 as the number of days. The results, including individual returns, drawdowns, wins, and sample observations, were saved. An inspection of the source confirmed that the entry was based on the same signal and close, with no fees or slippage, and included the current day, but did not validate the chronological order of the full window. Custom tests, numbered 79-82, were registered using vectorbt and used the next open and net costs, with daily and hourly training, and a separate exclusion of terminal marks and an independent cash ledger. However, no parameter was chosen from the command line interface. The positive control was unable to override the failed costs, small sample size, or absent live entry. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars and the engine is located at /app/agent/dist/backtest/engine.js.

A comparison of open interest on OKX USDT swaps from September 5, 02:11 to the baseline at 00:15 shows the following changes: USD open interest for Bitcoin decreased by 0.831%, for Ethereum increased by 0.241%, and for Solana increased by 0.122%. Funding forecasts are -0.115, 0.801, and -0.599 basis points, compared to the last settled rates of 0.007, 0.576, and -0.852 basis points. This analysis is based on three premises: first, changes in open interest on a single venue can be confusing because they combine price and contract quantity changes; second, previous data only included USD open interest, so the overall growth in contract numbers is unknown; and third, daily breakouts were false and the core verdict, known as verdict66, remains unchanged. Based on this, there is no conclusion of a leverage extreme or a squeeze entry. The current contract and coin quantities are now being observed for future comparison. A strong bull case may be supported by a negative Solana forecast, which could indicate that short sellers are covering their positions, while a bear case may be supported by persistent adverse expectations, but neither gives conditional payoffs. The data comes from the OKX API for funding rates and open interest. An existing research job, known as job13, remains in the research phase.

A nineteen-session runway simulation was conducted using NumPy stationary bootstrap with a seed of 90502, which sampled 10,000 paths of 19 trading sessions from each fixed strategy's net daily returns for 2023-Sep4, with a mean block of 20 sessions. This simulation represents a stylized near-month horizon and not an exact calendar or signal-conditioned forecast. The results showed that the range terminal returns were -2.77065%, +0.00468%, and +3.39568% for the 5th, 50th, and 95th percentiles, respectively, with a 40.63% loss probability and a 95th path drawdown of 3.66277%. A hypothetical full allocation of $245.69 would have resulted in returns of -$6.81, +$0.01, and +$8.34. The monthly version of the simulation showed returns of -6.49862%, +1.40311%, and +11.45624%, with a 38.07% loss probability and a 95th path drawdown of 11.37672%, resulting in dollar returns of -$15.97, +$3.45, and +$28.15. The actual book remains in cash. The simulation has limitations, including the potential for missing unseen tails and exact stop timing, and not calibrating the next entry. The range empirical p-value was 0.43609, with a W-value of 0.0249641 and an L-value of 0.0178879, resulting in a binary mean-payoff quarter-Kelly notional of 0.44756 and a monthly plug-in of 1.90339. However, these estimates are overconfident and not authorizations or risk fractions, and the range confidence spans zero and live exits remain unvalidated. The conservative eligible allocation is 0, with a maximum position of 212.77 and a nominal 1% equity of 8.5108, which remain ceilings and not targets. A funding need of $348.92 cannot alter these distributions. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the analysis was conducted using artifacts/stats.py and statistics analysis. The facts presented are based on measured historical paths, while the simulated probabilities are conditional model output.

A country momentum dependence audit was conducted. The audit analyzed 130 completed monthly trades and 133 range trades from September 4, 2023, to 2026. The trades were grouped by entry month, resulting in 44 and 45 clusters. Using NumPy with a seed of 90502, 10,000 independent month-cluster resamples were generated, producing mean-return 95% intervals of 0.21096% to 3.50982% for monthly trades and -0.56978% to 0.72463% for range trades. A stronger serial-dependence sensitivity test was also conducted with a seed of 90503, using stationary blocks that averaged three entry months, and 10,000 samples, resulting in mean-return 95% intervals of 0.35802% to 3.46067% for monthly trades and -0.51449% to 0.60859% for range trades. The positive fractions of 0.9938 and 0.6264 are bootstrap frequencies, not current win probabilities. Daily ordinary least squares analysis against the SPY ticker showed a monthly beta of 0.93818 and an annual arithmetic intercept of 3.00129%, with an R2 of 0.38703, while the range beta was 0.06893 and the annual arithmetic intercept was -0.53141%, with an R2 of 0.01884. The correlation between EWY and EWT was 0.76154, between EWY and EWP was 0.47974, and between EWT and EWP was 0.52041. The monthly annual returns were 14.83% for 2023, -8.26% for 2024, 40.23% for 2025, and 44.37% for 2026 to date, while the range returns were 3.25%, -2.08%, 6.00%, and -4.38% for the same periods. The conclusion is to distinguish between positive monthly historical evidence and failed range implementation, and not to reject all momentum or use monthly payoff to size range exits. The soundness of the audit is bounded by current surviving wrappers, one unusually strong recent chip cycle, historical rather than prospective holdout, and monthly dependence beyond resampling assumptions. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the scripts used are artifacts/stats.py and monthblock.py, with backtests 70-77.

The universe of stocks was created from the Nasdaq Trader lists nasdaqlisted.txt and otherlisted.txt, which were made on September 4 at 21:31, and included 13,155 unique securities, including ETFs, which stands for Exchange-Traded Funds. The list was filtered to remove non-standard symbols, which caused errors, and the remaining symbols were retried, resulting in 12,576 snapshot entries and 12,285 usable price pairs. A total of 540 non-standard symbols were omitted and 579 did not have snapshots, leaving 95.6% coverage of exchange-listed stocks, but not all US and over-the-counter stocks. The filters used were a price of $5 or more, a daily dollar volume of $100,000 or more on the IEX, which stands for Investors Exchange, and only stocks, not ETFs. This left 3,012 rows of data from September 4 and September 3. The one-day return was -3.132% at the 5th percentile, 0.0489% at the median, and 4.140% at the 95th percentile, with 1,543 positive returns. The original subset of 1,392 stocks was replaced and the comparison to stale PDEX, which stands for Previous Day's Exchange Data, was removed. The top positive and negative names were read across the whole universe, not just holdings, and included NX, AOUT, and biotechs, as well as AXTI, AEHR, and PSIX for primary supply exposure, and VSH, POWL, and ETN as fundamental comparators. No causal inference was made from the ranking. A total of 37 physical supply names were selected from Japan, Korea, Taiwan, Europe, Canada, Australia, and Hong Kong, but only 2454.TW had a prior 20-close breakout, although several series ended on September 3 and required a freshness check. This is targeted foreign coverage, not all foreign listings, and AUR.AX was not adopted due to being a speculative penny stock. A total of 8 cryptocurrencies were requested, but only 5 were returned, including BTC, ETH, SOL, FIL, and RENDER, and none were above their prior 20-close high. The missing cryptocurrencies FET, AR, and AKT were not proof that they were unavailable for trading, and no complete crypto census was claimed. The market data used was from a partial venue and Yahoo's indicative foreign history, not primary exchange quotes. The screening rule was exploratory and did not validate portfolio alpha or foreign execution eligibility. The sources used were https://www.nasdaqtrader.com/dynamic/SymDir/nasdaqlisted.txt and otherlisted.txt, https://runway.adityaperswal.workers.dev/data/stock-snapshots, https://query1.finance.yahoo.com/v8/finance/chart/, and https://runway.adityaperswal.workers.dev/data/crypto-bars. Every measured screen row and source code was preserved in 373 numbered compressed observation parts, with a hash in the note, and primary findings were individually recorded.

A trading fund manager used a computer program to test different trading strategies. The program, which used a mathematical technique called NumPy to generate 10,000 possible paths, or scenarios, with a seed value of 905, found that a circular strategy with 3-trade blocks, using 10% of changing equity per trade, resulted in different outcomes for each symbol. The natural exit count, or the number of times a trade was stopped, was 40, 39, 29, 38, 34, and 32 for each symbol. The program also modeled costs and found that the compounded results were: PSIX with 29 trades had a terminal value with a 5th percentile of +2.827%, a median of +11.277%, and a 95th percentile of +20.801%, with a loss frequency of 1.21% and a 95th percentile trade-step drawdown of 3.869%; POWL with 34 trades had a range of -2.060% to +7.829%, with a loss of 18.22%; AXTI with 40 trades had a range of -8.948% to +9.197%, with a loss of 52.57%; AEHR with 39 trades had a loss of 53.15%; and ETN with 32 trades had a loss of 93.78%. These results are based on empirical resampling frequencies, which means they are based on actual data, and do not take into account the probability of future trades or the dependence between different assets. The Quarter-Kelly capital plugin, a mathematical formula used to calculate optimal bet size, resulted in values of 181.55% for PSIX, 101.28% for POWL, 45.53% for VSH, and negative values for AXTI, AEHR, and ETN. However, the plugin's results are unstable and can produce absurd values. The trading strategy is also limited by a 25% cap on venue, but the eligibility criteria bind first, meaning that there are no positive signals, and PSIX has no training data. The actual risk is 0, and the reference ADR, or average daily range, is used to calculate the nominal notional risk ceilings, which are 73.45 for AXTI, 79.83 for AEHR, 212.77 for PSIX, 198.19 for VSH, 212.77 for POWL, and 212.77 for ETN. The start cash is 278.67, and the goal is to achieve repeatable net returns with an independent implementable edge. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, which provides exact statistics for all six archive stocks. A monthly shortfall of 348.92 does not alter the model.

A news scan found the following text: Premise 1 uses 419 complete common daily return rows from September 4, 2025, from all-adjusted IEX. This compares ordinary least squares, or OLS, versus SPY beta for several stocks: AXTI is 2.511, AEHR is 3.447, PSIX is 2.892, VSH is 2.387, POWL is 1.899, and ETN is 1.539. The correlation between POWL and ETN is 0.6593, and between AEHR and VSH is 0.5628. This shows that distinct products do not eliminate shared market exposure. The intercepts are descriptive, not causal alpha, and there is no annualized alpha claim. Premise 2 uses a mean of 10,000 circular three-trade block bootstrap seed 905 on natural exit samples of 40, 39, 29, 38, 34, and 32. Five means have a 95% interval crossing zero. The mean for PSIX is 3.7507%, with a confidence interval of 0.4832% to 7.1587%, but this is based on only four training bars, which is the selected post-listing history. An additional 20,000 draws give a conservative six-test tail sensitivity of -0.6271% to 8.3548%. This cannot correct all 3012-name screening and unknown prior research, and is not a formal population guarantee. Removing the best PSIX trade mean of 2.5700% shows that it is not entirely one outcome, and there is still small sample and regime risk. Premise 3 notes that contemporary survivorship, selected high movers, assumed costs, conditional barrier fills, and missing natural ledger losses undermine transportability. The conclusion is that PSIX merits prospective collection, not size, and others are rejected or watched per backtests 83-88. The source is https://runway.adityaperswal.workers.dev/data/stock-bars, with stats.json, tests.json, and derived.json/trades archived. The prior on PSIX shifts from unknown to interesting diagnostic, but was never adopted, and genuinely independent net execution outcomes would change the verdict.

A news scan found the following information about stock prices. On September 4, the IEX closes were AXTI at 61.62, AEHR at 86.245, PSIX at 40.54, VSH at 31.78, POWL at 181.22, and ETN at 410.805. All of these prices were below their prior 20-day close maxima of 96.025, 145.555, 43.05, 35.53, 218.9, and 459.85. PSIX was the only one of these six stocks that was above its 50-day Simple Moving Average, which is a calculation of the average price of a stock over a certain period of time. The one-day rises of AXTI and AEHR were 9.59% and 13.13%, but these stocks have had 20-session losses of 30.41% and 16.27%. This rebound is not a registered close-breakout, which is when a stock's price breaks out above a certain level, plus a 50-day Simple Moving Average conjunction, which is when the stock's price is above its 50-day average. The indicators used to analyze these stocks share price input, which means they are based on the stock's price, and are not independent confirmation. The 20-day Average Daily Range, which is a measure of the average daily price range of a stock, for each of these stocks is 7.14025, 9.19425, 1.50575, 1.36475, 6.8655, and 10.81175. These numbers provide a bounded reference move, which is a measure of the stock's price movement, but are not validated support, which means they are not a reliable indicator of the stock's price. The reference target for PSIX is 43.5515, which is two Average Daily Ranges from 40.54, but the prior 20-day high of 43.05 constrains the upside, and there is no actual entry, which means there is no signal to buy the stock. The whole-share affordability, which is the price at which a whole share can be bought, is under 212.77 for each of these stocks, but the risk limit, which is the maximum amount of money that can be lost, is not specified. One share of AXTI risks 7.14 at its nominal Average Daily Range, and one share of AEHR exceeds its equity risk of 8.5108. ETN requires a fractional share, which is a share that is less than one whole share, and this restricts extended hours exits, which are trades that are made outside of regular trading hours. There are no real positions to adjust, which means there are no actual trades to be made. The conclusion is that there is no entry signal, which means there is no signal to buy the stock, and negative, unstable, or missing training backtests, which are tests of a trading strategy, independently prohibit risk, which means they indicate that the trade is not worth the risk. Price and volume partial IEX, which is a stock exchange, and dividend and 15-minute enforcement gaps, which are gaps in the data, remain. The source of this information is https://runway.adityaperswal.workers.dev/data/stock-bars, and the technical.json and per-symbol observations are archived.

The primary issuer observations were saved on September 5. The iShares fact sheets from June 30 show that EWY, which tracks the South Korean market, has a significant weighting of 50.36% in Hynix and Samsung, while EWT, which tracks the Italian market, has a weighting of 75.11% in a single sector, and EWP, which tracks the Philippine market, has weightings of 43.70% in financials and 24.16% in utilities. The price-to-earnings ratios of 27, 32.40, and 15.57 reflect different sectors and issuer loss-exclusion or capping, not comparable discounts. A news scan found the fact sheets at https://www.ishares.com/us/literature/fact-sheet/ewy-ishares-msci-south-korea-etf-fund-fact-sheet-en-us.pdf, and similar fact sheets for EWT and EWP can be found with the same filename pattern. The premise is that country labels do not eliminate shared chip cycle or valuation risk. On September 3, NX released pricing-led sales of +1.3%, debt paydown of $42.25 million, but 9-month EBITDA of $144.3 million versus $172 million, with a GAAP bounce affected by prior impairment, as found at https://investors.quanex.com/news-releases/news-release-details/quanex-building-products-announces-third-quarter-2026-results. On July 14, AEHR reported FY revenue of $50 million, GAAP net loss of $7.1 million, and on August 12, received a $22 million order, which is 44% of annual sales, with customer concentration and delivery conversion mattering, and forecast FY 2027 revenue of $130-150 million, as found at https://www.aehr.com/2026/08/aehr-receives-22-million-follow-on-order-for-ai-processor-wafer-level-burn-in-systems/. On August 5, SNDK reported a quarter gross margin of 84.6% and adjusted EPS of $39.25, which annualizes to a peak quarter, creating an 11.08x multiple versus FY adjusted 24.55x at $1740, but neither proves fair value, as found at https://investor.sandisk.com/news-releases/news-release-details/sandisk-reports-fiscal-fourth-quarter-2026-financial-results. There is a strong bull case for earnings and order growth, and a profitable country holdout, but also a bear case for embedded expectations, sector crowding, partial cash conversion, and cyclical normalization. The conclusion is to watch research only, as no fresh issuer fact supplies independently validated conditional payoff. Issuers are closest to metrics but benefit from positive framing, with no independent consensus-surprise model, fulfilling valuation, or future growth guarantee claimed.

A news scan found the following information: on September 4 at 21:31, a review of Nasdaq and other listed directories showed 13,155 non-test instruments, with 5,190 non-standard symbols excluded, leaving 12,636 queried and 12,305 with valid daily or prior bars. A filter was applied to find instruments with a price of $5 or more and a daily volume of $1 million or more, which resulted in 2,380 instruments, including 1,908 non-ETFs and 472 ETFs, with 47.18% having a positive day and a median return of -0.0411%. The 5th and 95th percentiles were -3.2416% and +4.1017%, respectively. This data is based on IEX daily volume, which is not consolidated volume. Further analysis of 113 selected non-ETFs found 109 full 12-1 signals, but today's selection cannot define historical out-of-sample membership. The review also included 19 non-US country wrappers and 14 Yahoo local series, extending the reach but not providing an exhaustive world census. A crypto screen returned 29 histories out of 33 USD requests, with ALGO, APE, ATOM, and NEAR missing, but not proven untradable. USDC and USDT were excluded from risk momentum due to unvalidated volume anomalies. BTC, ETH, and SOL had 12-1 negative returns of -41.95%, -57.65%, and -63.74%, respectively, despite 20-day returns of +26.42%, +30.57%, and +35.42%, illustrating different horizons. PAXG had insufficient formation. The conclusion is that the research candidate pool has been expanded, but no entry authorization has been given. The screen leaders were daily NX +21.18% and AEHR +13.13%, with 12-1 SNDK +2215%, MU +593%, and WDC +580%. Geared SNDK ETFs were excluded from cash strategy replication. Country EWY, EWT, and EWP led, with local SAP having a 20-day return of +8.54%, but the latest data was from September 3, not September 4. There are explicit gaps in the data, including missing preferred and warrant conventions, OTC, delisted point-in-time universes, complete crypto list, and global census. The data was saved in a raw all-scan CSV with 175 detailed history observations. The sources of the data include Nasdaq Trader, Runway Aditya Perswal, and Yahoo Finance. Prices are measured as past returns, not fundamental forecasts. The original paper distinguishes cross-sectional momentum from one-day headline chasing.

A news scan found the text: Numpyseed2026090502; each symbol independently draw three complete 24h log return blocks from 2026 history, 10k 72h paths. XRP/LINK/UNI 245 usable day blocks, PAXG 193; previous close-to-current close hour returns; no unseen intra-hour movement or 15min venue checks. Reference Sep 4 close with 1 ADR 20 stop, 2 ADR 20 target widths respectively 7.609/15.218%, 5.912/11.823%, 7.126/14.251%, 3.685/7.370%. Observed hourly close target before stop frequencies 3.71/5.67/8.51/.60%; stop first 9.53/22.25/24.30/13.93%; neither 86.76/72.08/67.19/85.47%. Unconditional resampling, not actual stop-touch probabilities, signal-conditioned forecast or weekend-specific sample. 72h terminal 5th/95th XRP -7.764/+10.303%, LINK -8.471/+10.095%, UNI -10.435/+13.670%, PAXG -4.236/+4.179%. A measured 2 ADR target can still be unlikely inside 3 days; do not stretch holding window merely to fit. Nominal 1% equity risk 8.5108 yields notional ceilings 111.86/143.97/119.44/212.77, not guaranteed loss limits. Actual size 0 due uncertain net edge. Validity supports caution about attainability; soundness limited by resampling past blocks, separate assets, stationarity and close only paths. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; artifacts/stat_risk.py/stats-risk.json.

A data analysis from the source https://runway.adityaperswal.workers.dev/data/crypto-bars found that from January 1 to September 4, 2026, the net natural-exit samples for the cryptocurrencies XRP, LINK, UNI, and PAXG were 6, 14, 13, and 8, respectively. Using 20,000 independent and identically distributed resamples with a seed of 2026090502, the mean trade 95% intervals were calculated as XRP [-1.194, 18.765]%, LINK [-1.726, 2.184]%, UNI [-2.933, 2.634]%, and PAXG [-1.402, 0.419]%. This analysis assumes that the trades are independent, but the current selection of trades and market conditions may limit the interpretation of the results. The frequencies of positive resamples were 94.375% for XRP, 51.585% for LINK, 38.65% for UNI, and 13.8% for PAXG, but these are not posterior probabilities. A daily ordinary least squares analysis on BTC with an intercept found betas of 1.237, 1.176, 1.181, and 0.299 for XRP, LINK, UNI, and PAXG, respectively, with correlations of 0.871, 0.874, 0.677, and 0.446. The sample size was 247 for each cryptocurrency except PAXG, which had a sample size of 204. The similar betas for XRP and LINK weaken the narrative of diversification. The raw binary mean-payoff quarter Kelly notional formula gave capital fractions of 3.889, 0.281, -0.468, and -7.550, which are clearly unstable and constrained before caps. Since the lower mean bounds are negative, the authorized allocation is 0. The premises of a positive XRP return and a win rate of 66.67% depend on only 6 trades, and the strongest bull case is a large payoff, but no precise population edge has been established. There is a need for prospective fixed rule evidence and tested venue exits, and one should not pick the best lookback from the inspected holdout. The UNI terminal open mark was excluded from payoffs rather than counted as a winner.

On September 5 at 02:00, research was conducted. It involved looking at 13,155 securities listed on NasdaqTrader, which includes 5,584 Nasdaq securities and 7,571 securities from other exchanges, as well as 7,499 non-ETF rows including non-common instruments. These were cross-checked with 19 weekday Nasdaq earnings-calendar requests from September 8 to October 2. The result was 214 unique listed candidates, but 122 of these calendar rows did not have a release time. It's important to note that these are estimates and not confirmations from the issuers. Daily bars from IEX were fetched for all 214 candidates, plus six comparison and regulatory names, and 218 out of 220 had at least 51 observations. A filter was applied where the price is greater than the 20-day simple moving average, which is greater than the 50-day simple moving average, and the close is less than or equal to 49.14, leaving 40 candidates. These were then ranked by observed IEX dollar volume, with CPRT being the first. This process is a screening and not a validated strategy or a complete current common-stock history. International RTT pages returned 419 rows, but 99 were rejected because their displayed date differed from the requested day, leaving 320 date-matched rows, of which 254 were classified as foreign due to their suffix. Only the first-page coverage and six selected foreign quotes were considered, and this was not an exhaustive global price census. The Coinbase 409-currency catalog was not used as a venue universe or a dated catalyst list because protocol upgrades are outside this fund's earnings and FDA mandate. The ETH date review belongs to the existing Crypto monitor. The primary-source work, not secondary dates, determines eligibility. The validity of the candidates can be generated without accepting estimated dates, but the soundness is limited by directory survivorship, missed calendar companies, pagination, and partial IEX liquidity. The names that were rejected due to missing dates remain ineligible and are not considered bearish. The sources used for this research include https://www.nasdaqtrader.com/dynamic/SymDir/nasdaqlisted.txt, https://www.nasdaqtrader.com/dynamic/SymDir/otherlisted.txt, https://api.nasdaq.com/api/calendar/earnings?date=2026-09-09, https://www.rttnews.com/Calendar/Earnings.aspx?Date=08-Sep-2026, https://api.exchange.coinbase.com/currencies, and https://runway.adityaperswal.workers.dev/data/stock-bars.

This is a description of an automated trading strategy for an Exchange-Traded Note (ETN). The strategy is based on three premises: (1) it uses a fixed trend-plus-breakout approach with next-open range exits, where the protocol note127 fixes the prior 20-close breakout and Simple Moving Average 50 (SMA50), with a next-open entry and a frozen signal-day Average Daily Range 20 (ADR20) stop and target, and a timeout after 5 sessions at the next open, with a long-only position. (2) The strategy uses vectorbt to represent two events per daily bar, which are the open and the later barrier or close, including entry-day exits, with synthetic timestamps that do not claim intraday observations. Adverse gaps fill the open, while favorable target gaps conservatively target, with dual touches stopping at the first touch. Fees are 5 basis points and slippage is 10 basis points on each side, with stress doubling both. Independent cash and share values reconcile with a maximum error of 3.637978807091713e-12. (3) The strategy was trained from 2022 to 2024 and diagnosed from September 4, 2025, onwards, using all-adjusted IEX data, which is paginated. The strategy rejects training and later-period net returns that are negative. A training sweep of targets and stops resulted in the following returns: (1.5, 1) = -12.752220910464256, (1.5, 1.25) = -6.8086762427363094, (2, 1) = -10.708581864755178, and (2, 1.25) = -12.052126108536413. The strategy notes that all contemporary selected names have survivorship and selection bias, and no pristine prospective evidence is claimed. Terminal open marks are excluded from the payoff sample, with currentholdoutopen0 and trainingopen0. An initial entry-day omission was detected and corrected before journal metrics, and discarded output was not used. The strategy does not use cash dividends, 15-minute monitoring, partial fills, or whole-share execution models. Daily target touches may not execute under venue timing, which means the strategy is not live eligible, regardless of favorable historical results. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the executable test.py and full trades are stored in a measured-dataset archive. The current signal is false.

The POWL trading strategy is based on three premises. First, it uses a fixed trend-plus-breakout approach with next-open range exits, where the prior 20-day close breakout and 50-day simple moving average (SMA50) are considered, and the next open is used as the entry point, with a frozen signal-day average daily range (ADR20) stop and target. Second, the strategy uses a vectorbt approach with two event rows per daily bar, representing open and later barrier or close, and synthetic timestamps that do not claim intraday observations. Adverse gaps are filled at the open, and favorable target gaps are targeted conservatively, with dual touches stopping at the first touch. Fees of 5 basis points and slippage of 10 basis points are applied to each side, and stress tests double both. Independent cash and share values are reconciled with a maximum error of 3.637978807091713e-12. Third, the strategy was trained from 2022 to 2024 and diagnosed from September 4, 2025, onwards, using all-adjusted IEX data. A watchlist of positive selected later periods shows that all four training variants lose, and the mean interval crosses zero. The training sweep targets 1.5 and 2, with stops at 1 and 1.25, resulting in losses of -6.741325252297993, -18.69843938469283, -3.852527187305641, and -21.788176286351028, respectively. The strategy has survivorship and selection bias, and no pristine prospective evidence is claimed. Terminal open marks are excluded from the payoff sample, and the current holdout open is 0, with the training open also at 0. An initial entry-day omission was detected and corrected before journal metrics, and discarded output was not used. The strategy does not account for cash dividends, 15-minute monitoring, partial fills, or whole-share execution models. Daily target touches may not execute under venue timing, making the strategy not live eligible, regardless of favorable historical results. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the executable test.py and full trades are stored in the measured-dataset archive. The current signal is false.

A trading strategy for VSH uses a fixed trend-plus-breakout approach with next-open range exits. The strategy has three main premises: (1) it uses a protocol that notes the prior 20-day close breakout and the 50-day simple moving average, with next-open entry and a frozen signal-day average daily range stop and target, and a timeout after five sessions at the next open, only for long positions. (2) The strategy uses vectorbt to represent open and close barriers, including entry-day exits, with synthetic timestamps that do not claim intraday observations. Adverse gaps fill the open, and favorable target gaps conservatively target, with dual touches stopping at the first touch. Fees are 5 basis points and slippage is 10 basis points on each side, with stress doubling both. Independent cash and share values reconcile with a maximum error of 3.637978807091713e-12. (3) The strategy was trained from 2022 to 2024 and diagnosed from September 4, 2025, using all-adjusted IEX data. However, all four training variants lost, and double costs turned the holdout negative. The training sweep targeted stops of 1.5 and 2, with results of -13.635595520075913, -19.17348293114387, -3.52923149049484, and -8.73000753401997 for the respective stop and target combinations. The strategy has survivorship and selection bias, and no prospective evidence is claimed. The current signal is false, and the strategy is not eligible for live trading, despite favorable historical results. The data source is https://runway.adityaperswal.workers.dev/data/stock-bars, and the executable test and full trades are stored in the measured-dataset archive.

A trading strategy for PSIX uses a fixed trend-plus-breakout approach with next-open range exits. The strategy has three main premises: (1) it uses a protocol that combines a prior 20-day close breakout and a 50-day simple moving average, with a next-open entry and a frozen signal-day average daily range stop and target, and a timeout after five sessions at the next open, only for long positions. A simple moving average is a calculation of the average price of a stock over a certain period. (2) The strategy uses a vectorbt approach, which is a backtesting framework, with two event rows per daily bar representing open and later barrier or close, including entry-day exits, and synthetic timestamps that do not claim intraday observations. Adverse gaps fill open, and favorable target gaps conservatively target, with dual touches stopping at the first touch. The strategy also includes fees of 5 basis points and slippage of 10 basis points on each side, with stress tests that double both. (3) The strategy was trained on data from 2022 to 2024 and diagnosed on data from later periods, up to September 4, 2025, using all-adjusted IEX data, which is a type of stock exchange data. The training sweep targets stops of 1, 1.25, 1.5, and 2, with results of 0 for each combination. The strategy has several limitations, including survivorship and selection bias, and does not claim to have pristine prospective evidence. Terminal open marks are excluded from the payoff sample, and the current holdout open is 0, as is the training open. The strategy does not use cash dividends, 15-minute monitoring, partial fills, or whole-share execution models. Daily target touches may not execute under venue timing, so the strategy is not eligible for live trading, regardless of favorable historical results. A news scan found the source of the data to be https://runway.adityaperswal.workers.dev/data/stock-bars, with an executable test.py and full trades stored in a measured-dataset archive. The current signal is false.

The automated trading fund manager has a strategy for the stock AEHR that is based on three premises. First, the strategy uses a fixed trend-plus-breakout approach, which means it looks at the stock's prior 20-day closing price and its 50-day simple moving average, or SMA50, to decide when to enter a trade at the next open. The strategy also uses a stop-loss order, which is set at one times the average daily range, or ADR20, and a target price, which is set at two times the ADR20. If the trade is not successful after five sessions, it is closed at the next open. This strategy only considers long positions, meaning it only buys the stock. Second, the strategy uses a system called vectorbt, which represents each day as two events: the open and the close. This system does not claim to have intraday observations, meaning it does not look at the stock's price during the day, only at the open and close. The strategy also takes into account gaps in the stock's price, both adverse and favorable, and uses a dual touch approach to set stop-loss orders. The strategy includes fees of 5 basis points and slippage of 10 basis points on each side of the trade, and these fees and slippage are doubled in stressful market conditions. The cash and share values are reconciled to ensure accuracy, with a maximum error of 3.637978807091713e-12. Third, the strategy was trained on data from 2022 to 2024 and was tested on data from 2025 up to September 4. The strategy uses adjusted data from the IEX exchange and is paginated, meaning it is divided into pages or sections. However, the strategy was rejected because its net returns were negative during both the training and later periods. The strategy was tested with different target and stop-loss prices, and the results were: (1.5, 1) -38.643851939335995, (1.5, 1.25) -57.67302561057384, (2, 1) -32.86874570103676, and (2, 1.25) -46.090402422060095. The strategy notes that all contemporary selected names have survivorship and selection bias, meaning they have been chosen because they have performed well in the past, and there is no pristine prospective evidence to support the strategy. The strategy also notes that terminal open marks are excluded from the payoff sample, and the current holdout open is 0, while the training open is 1. An initial entry-day omission was detected and corrected before journal metrics were calculated, and discarded output was not used. The strategy does not use cash dividends, 15-minute monitoring, partial fills, or whole-share execution models. Daily target touches may not execute under venue timing, which means the strategy may not be eligible for live trading, regardless of its favorable historical results. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the executable test.py and full trades are stored in a measured-dataset archive. The current signal for the strategy is false, meaning it is not recommending a trade at this time.

A trading strategy for AXTI uses a fixed trend-plus-breakout approach with next-open range exits. The strategy has three premises: (1) it uses a protocol note127 with a prior 20-day close breakout and a 50-day simple moving average, with a next-open entry and a frozen signal-day average daily range stop and target, and a timeout after five sessions at the next open, only for long positions. (2) The strategy uses vectorbt with two event rows per daily bar to represent open and later barrier or close, including entry-day exits, and synthetic timestamps do not claim intraday observations. Adverse gaps fill the open, and favorable target gaps conservatively target, with dual touches stopping at the first touch. Fees are 5 basis points and slippage is 10 basis points on each side, with stress tests doubling both. Independent cash and share values reconcile with a maximum error of 3.637978807091713e-12. (3) The strategy was trained from 2022 to 2024 and diagnosed from September 4, 2025, onwards, using all-adjusted IEX data, paginated. The strategy rejects training and later-period net returns that are negative. A training sweep of targets and stops resulted in the following net returns: (1.5, 1) -31.442725836626114, (1.5, 1.25) -35.08498901772668, (2, 1) -34.23281713352532, and (2, 1.25) -38.59042367158891. The strategy notes that all contemporary selected names have survivorship and selection bias, and no pristine prospective evidence is claimed. Terminal open marks are excluded from the payoff sample, and the current holdout open is 0, with the training open also 0. An initial entry-day omission was detected and corrected before journal metrics, and discarded output was not used. The strategy does not use cash dividends, 15-minute monitoring, partial fills, or whole-share execution models. Daily target touches may not execute under venue timing, so the strategy is not live eligible, regardless of favorable historical results. The source of the data is https://runway.adityaperswal.workers.dev/data/stock-bars, and the executable test.py and full trades are stored in a measured-dataset archive. The current signal is false.