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.
16,761 documents
The article describes a Chinese equity screening rule for Shenzhen Main Board stocks. It selects companies with turnover between 3% and 12%, a 2021 listing year, price-to-earnings ratios from 0 to 29.01, and price-to-book ratios from 0 to 3.11. The rationale…
This study constructs equity selection factors from trade by trade data to measure aggressive buying. It aggregates transactions into minute level buying and selling amounts, then derives buying share and buying intensity measures. The proposed factors can…
The document describes a rules-based gold strategy that buys after the CME’s daily maintenance break, holds for a fixed period, and uses a volatility-scaled stop. It reports an eleven-year hourly-data study: the first hour after the daily reopen showed a…
The document explains how to run strategy backtests in OctoBot: choose a profile, download historical market data, select a data file, run the simulation, and review performance summaries, charts, and individual trades. It describes how the starting…
The Expert Advisor uses the LSMA Angle histogram to generate signals at bar close. It opens a position when the histogram breaks through an overbought or oversold level, then closes the position when the histogram crosses the zero level. This defines a…
This Expert Advisor uses signals from the ColorJJRSX oscillator. A signal forms when a bar closes and the indicator's direction and color have changed. The EA therefore relies on a completed bar for its entry signal rather than reacting to an intrabar color…
This configuration specifies a Qlib workflow for training a LightGBM model on the Alpha158 feature set for China’s CSI 300 universe. It defines training, validation, and test periods, along with model settings such as mean squared error loss, tree depth,…
The document shows a Pine Script strategy that enters long when a short-period RSI crosses below a threshold, then closes when price rises above an exponential moving average or falls to a fixed percentage below the recorded entry price. The author’s…
The article describes a framework for identifying crowded trading and distinguishing a bubble’s buildup from its unwinding. It estimates asset clustering from covariance and principal components, using the concentration of sector returns in important…
This study proposes a passive index that selects between the A-share and H-share of the same company according to which listing is cheaper after currency adjustment. It examines price gaps across dual-listed Chinese companies, using panel regression to…
This community post proposes a screen for Chinese equities combining a market capitalization below 10 billion yuan, no reported past losses, a positive intraday increase in holdings above 5%, and an auction-stage price change between negative 2% and positive…
This research note evaluates 31 quality factors for stock selection in Hong Kong equities and compares their effectiveness with results in mainland A-shares. Twelve factors are reported as notably effective, particularly measures tied to operating cash flow…
This Chinese-language post describes a stock selection screen based on three market conditions: daily price amplitude above a threshold, turnover between specified bounds, and a weekly bullish signal. The rationale is that more active, volatile stocks may…
This document describes a multiple-timeframe moving average indicator that reads values from the H4 and H1 timeframes, allowing a trader to view those readings while using a different chart timeframe. Its stated purpose is to retrieve an indicator value from…
This overview introduces statistical learning methods for prediction and data analysis. It covers linear and logistic regression, linear and quadratic discriminant analysis, resampling, predictor selection, regularization, dimension reduction, nonlinear…
This brief indicator description presents Kalman-filter velocity as a smoothed measure of price momentum. Its multi-timeframe version adds three higher-timeframe views beyond the standard chart intervals available in MetaTrader 5: the first, second, and…
The document introduces Altman's Z-score as a composite financial-risk measure built from five ratios: working capital, retained earnings, earnings before interest and taxes, market equity relative to liabilities, and sales, each scaled against assets except…
This research review examines momentum in China’s A-share market as institutional ownership and analyst attention have grown. It compares traditional lookback returns with measures that account for price paths, remove limit-up days, use earnings announcement…
This research note describes extending mean-variance portfolio allocation with higher-moment information, especially co-skewness and co-kurtosis among assets. Mean and covariance alone can miss differences in joint tail behavior: portfolios with similar…
This post proposes ranking Chinese equities by capital-flow strength while filtering for company scale and repeated limit-up sessions within a recent period. The rationale is that stronger flow may indicate investor attention, larger companies may be more…
This post outlines a Chinese equity screen that combines an amplitude threshold, a proxy for institutional buying, and a recent large daily gain. The intended logic is to find volatile stocks attracting institutional interest that have also shown a strong…
This forum post reports a failure during VeighNa’s genetic-algorithm optimization for a strategy backtest. The traceback reaches Python’s multiprocessing pool while it tries to send evaluation work to worker processes. Serialization then fails because the…
This Chinese stock-screening article combines a positive MACD reading with seven consecutive falling closes and a shortening MACD histogram on a 15-minute chart. The proposed interpretation is that shrinking negative histogram bars may signal weakening…
This Python utility converts Bybit historical depth and trade files into the event array format used by HftBacktest. It reads order book updates from a zipped JSON stream and trades from a gzip-compressed CSV, creates depth, snapshot, clear, and trade…