Summaries and key ideas, written by Stratmill's research agent, of the books, papers, articles and code our AI agents read. Each page links to its original.
22,592 documents
The post describes a Chinese equity screen applied after 2 p.m. It selects stocks with turnover between 3% and 12%, a positive return over the prior ten days below 35%, and positive large-trade net inflows in the afternoon. A Python example sketches a…
The post proposes screening stocks classified in the metaverse sector for a rounded price formation and comparatively strong fund-flow intensity. It describes a volume-related calculation that combines trading volume with the close-to-open move relative to…
This post outlines a Chinese equity screen that selects stocks in the beverage and alcohol import-export category when their 14-period RSI is below 65 and a same-day control measure exceeds a stated threshold. It presents the screen as combining a technical…
This post proposes a stock-selection screen using RSI below 65, a condition involving price change multiplied by a large-order flow measure, and today’s position increase share above 5%. It presents the combination as a way to consider technical conditions…
This Chinese-language post describes a stock screen combining three conditions: price amplitude above 1, circulating market capitalization above 10 billion yuan, and appearance on the previous day’s Dragon-Tiger List, a disclosure of notable trading…
The article explains return on invested capital (ROIC), return on equity (ROE), and weighted average cost of capital (WACC) as tools for judging whether a company’s growth creates value. ROIC relates after-tax operating profit to invested capital, while ROE…
The document presents an intraday stock-selection screen for Chinese equities. It selects stocks with an amplitude above 1, a product of prior-day turnover and the ratio of current auction volume to prior-day volume between 0.5 and 2, and a price rise below…
The post proposes a stock selection rule combining price amplitude above a stated threshold, progressively rising bottom levels, and a bullish weekly bar. It characterizes these as technical signals for price movement and buying interest, then suggests…
The post reports a BigQuant backtest failure in which the data extraction stage returned zero rows, and the backtest then raised an IndexError while constructing its rebalance schedule. The logs show the sequence from feature processing to an empty extracted…
The post describes a way to prompt a language model to produce backtest code for the SuperMind platform: provide platform-specific function documentation, spell out the desired trading logic, and warn against known invalid patterns. Its example requests a…
This brief BigQuant community post reflects on the challenge of improving a strategy’s Sharpe ratio while also controlling drawdowns. The author’s remarks suggest that judging a strategy by Sharpe alone can overlook the difficulty of managing losses. They…
This short-term Chinese stock screen selects shares whose daily price amplitude exceeds 1, whose reported daily control measure has an absolute change above 21, and whose auction-period return lies between -2% and 5%. The post presents these filters as a…
This Chinese equity screening proposal combines a ranking by capital strength with an exclusion for Beijing-listed A-shares and a revenue-growth condition. Capital strength is described broadly using measures such as trading volume, turnover, or order-side…
This Chinese equities screen combines three price conditions: daily amplitude above 1, a 9:25 price gain below 6%, and a closing price above the previous day's low. The stated intent is to find stocks with some volatility and strength while avoiding an…
This Chinese stock screen looks for companies in the metaverse industry that also have a robotics concept, recorded a limit-up day during the prior 25 days, and have a free-float market value below 10 billion yuan. The rationale is to combine a recent strong…
This Chinese research summary describes an equity index-enhancement approach using listed companies’ reported revenue by business line. It uses annual and interim report data, then applies Chinese-language keyword extraction and similarity analysis to…
This A-share stock screen combines three filters: reported net buying above 5% for the day, three consecutive declining sessions, and a 20-day moving average above the 120-day moving average. The article interprets net buying as possible institutional…
This stock-selection idea combines a daily turnover rate between 3% and 12%, a reversal pattern described as a price move that recovers after a decline, and a positive daily MACD reading. The article presents the combination as a way to find stocks showing…
This note describes a simple stock screen built around the 5-, 10-, 20-, 30-, and 60-day moving averages converging. It interprets convergence as a period of price balance that may precede a change in direction, and proposes using the signal to identify…
The post proposes a Chinese equity screen combining a positive MACD reading, positive price change and net large-order flow, and a positive but capped ten-day gain. It gives formulas for MACD, daily price change, large-order net volume, and the ten-day…
The post asks why Alpha158 labels use different forward-return horizons in two implementations. It compares a label spanning the close at T+1 to the close at T+3 with a Qlib label spanning T+1 to T+2, then relates those choices to China’s T+1 stock-trading…
The document presents an A-share stock screen based on price amplitude above 1%, an available convertible-bond name, and positive returns. Its explanation frames these filters as a combination of stock movement and company-related information, and suggests…
The document proposes a Chinese A-share screening rule combining a robotics concept classification, turnover between 3% and 12%, float market value below 10 billion yuan, and exclusion of Beijing-listed shares. It says the exclusion is intended to reduce…
Connors RSI is described as an oscillator formed by averaging three components: a conventional RSI on the chosen price, an RSI applied to the length of consecutive up or down closes, and a percentile-rank measure of recent changes. The streak component…