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
This Chinese equity screening idea selects non-ST stocks with turnover between 3% and 12% after seven consecutive declining sessions, with selection intended before 10 a.m. It describes a contrarian premise: a prolonged decline may create a rebound…
This Chinese equity screening note combines an RSI threshold with an external-to-internal trading volume ratio and a morning star pattern signal. Its prose gives an RSI ceiling of 65, a volume ratio above 1.3, and the pattern condition as criteria for…
This forum post describes a Chinese equity backtest comparison between standard regression models and a platform stock ranking model. The author trained XGBoost, LightGBM, and CatBoost regressors on five-day returns using rolling historical periods, then…
This note describes a Chinese equity screen combining daily price movement with a moving average trend filter. Its stated rules select stocks with an amplitude above 1, a daily maximum decline between 4% and 5%, and a 20-day moving average above the 120-day…
This post describes a Chinese-market stock screen combining five conditions: at least five moving averages converge, current volume exceeds 10,000 lots, the stock opens higher, and the 15-minute MACD histogram’s negative bars are getting shorter. The author…
This stock-screening rule selects shares with turnover between 3% and 12%, a rising K value in the KDJ technical indicator, and prior-day trading value above 60 million. The post frames the conditions as a combination of liquidity and a short-term technical…
This Chinese stock-selection proposal filters for shares with turnover between 3% and 12%, a positive daily price change, and a negative product of that price change and the net quantity attributed to very large orders. It also limits the universe to…
This proposed screen selects mainland Chinese main-board stocks using three short-term conditions: amplitude above a threshold, a rising and diverging moving-average pattern, and a buy-side volume ratio above 1.3. The article interprets these as signals of…
This indicator colors part of each price bar according to the candle’s direction. For a rising candle, it marks the segment from the open to the high in light green; for a falling candle, it marks the segment from the open to the low in pink. The display…
This research summary describes a multi-factor tracking framework built around the pricing logic of factors. It examines historical returns for technical, expectation-based, and financial factors using a modified cross-sectional regression approach. To…
This document describes a candlestick indicator that marks hammer-shaped and inverted-hammer-shaped candles. It classifies the two forms by the position of the candle body within its range: the body is in the upper or lower half. The indicator uses blue for…
This Chinese-language post proposes a stock screen requiring MACD to be above zero, a descending ranking of capital-flow strength, and a rising 30-day moving average. The stated intent combines a positive trend filter with a measure of fund activity. It…
This Chinese-language post outlines a short-term stock screen using three conditions: daily amplitude above 1, turnover between 2% and 9%, and a shrinking negative MACD histogram on a 15-minute timeframe. The intended rationale is to combine price movement…
The document presents a Chinese equity screening rule combining three conditions: prior-session price amplitude above 1%, appearance on the previous day’s trading leaderboard, and at least one daily gain of 10% or more within the previous 25 sessions. It…
This proposed Chinese equity screen combines a price-amplitude condition, simultaneous moving-average crossovers, and appearance on the prior day’s trading leaderboard. The technical filters are intended to identify active stocks with upward momentum, while…
This stock screen selects Shanghai-listed shares whose daily high-low range exceeds a threshold and that have reached the daily price limit at least twice over a recent 500-trading-day window. It combines a measure of price movement with repeated limit-up…
This technical screen looks for stocks with amplitude above 1, at least two limit-up events within the previous 500 days, and declines across seven consecutive sessions. The proposed interpretation is that the screen identifies weak recent price action in…
This Chinese stock screen combines a daily price-range condition, a short-term trend filter, and valuation limits. It seeks Shenzhen Main Board stocks with amplitude above 1, a price above the five-day moving average, a price-to-earnings ratio between 0 and…
This Chinese A-share screening idea combines a price-range condition with a volume and turnover measure. It selects stocks whose amplitude exceeds 1 and whose previous-day turnover multiplied by today’s auction volume divided by yesterday’s volume falls…
The proposed equity screen combines three conditions: rank stocks by trading volume and turnover value as a proxy for capital interest, restrict candidates to codes beginning with 60, and look for a shortening MACD histogram bar on a 15-minute chart. The…
This post presents an A-share stock screen that combines a daily price-range condition, market capitalization below the stated limit, positive net profit, and a buy-to-sell volume imbalance above 1.3. It includes an indicator-formula example and a Python…
This post describes a Chinese A-share screening rule combining a daily high-low range above 1%, an available convertible-bond name, and a positive return over ten days that remains below 35%. It offers an indicator-formula example and a Python example…
This Chinese-language post outlines an A-share stock-selection screen combining a positive MACD condition, rising or diverging moving averages, market capitalization below 10 billion yuan, and positive earnings per share. It also suggests ranking candidates…
This BigQuant example addresses a limitation in which its high-frequency feature extraction module calculates within a single day, while a researcher may need features from minute data spanning multiple days. It shows a distributed workflow using FAI: read…