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
This tutorial shows how to build a long-only moving average crossover strategy by combining reusable strategy components from a Python backtesting library. It turns the relationship between a short and a longer moving average into entry signals, allocates…
This meetup page collects questions about quantitative trading on the BigQuant platform. Topics include searching for holding-period parameters in a default stock-ranking template, defining reusable Python modules, and building a workflow for developing…
This configuration describes a Qlib machine-learning workflow that trains a CatBoost regression model on Alpha158 features for CSI 300 instruments. It defines separate training, validation, and test periods, then records signal analysis and portfolio…
This short description outlines an expert advisor that trades Aroon indicator crossovers on a 15-minute chart. Aroon up and down trend variables can be used to confirm the direction on an hourly chart. A Williams Percent Range oscillator is added to help…
This research summary describes factors derived from operating financial statements and reports selected long-short results. It identifies changes in operating current liabilities as a notable factor, with a reported Sharpe ratio of 2.62 and annualized…
The document describes a script that compares streams of price bars to find a similar historical sequence and illustrates the resulting match alongside a predicted bar and price area. Inputs control the comparison-window length, the number of bars shown, and…
The document presents a Chinese equity screening rule that selects stocks with RSI below 65, circulating market value between 5 billion and 10 billion yuan, and turnover between 3% and 12%. It frames RSI as a short-term overbought or oversold measure, market…
This research summary proposes stock-selection factors built from daily highs, lows, opens, and average traded prices, arguing that closing-price indicators alone miss information in price movement. It evaluates opening-price spikes, rebounds from intraday…
This stock-selection rule combines three conditions: MACD must be above zero, the share price must be below 12 yuan, and at least one daily gain of 10% or more must have occurred within the prior 25 trading days. The post gives both an indicator-style…
This document outlines an Expert Advisor that trades when the Ozymandias indicator’s middle line changes color. It is a simple signal-based approach, with the color transition serving as the trade decision trigger. The system requires the compiled indicator…
The document describes a method for testing factor effectiveness dynamically and selecting stocks within industries. It examines whether differences in style-factor exposure relate to differences in stock returns, then uses the results to form industry-based…
This overview explains the main stages of a machine-learning workflow for quantitative investing, using a fruit-selection analogy to introduce training data, labels, features, prediction, and validation. It recommends defining the market and stock universe,…
This Qlib demonstration explains how to reuse a processed data handler across repeated model training runs. It first trains the same configured task more than once without explicitly reusing the handler, then constructs the configured data handler in memory…
The document describes a simple A-share stock screen using a minimum daily high-low range, a specified closing price, and a bounded daily return. It frames the filters as a way to select for price movement while restricting the current price and recent…
This research summary examines stock selection factors derived from operating financial statement items, especially changes in operating current liabilities. It reports that these factors showed selection ability, with the strongest cited result for a…
This research overview examines risk parity within the broader development of portfolio allocation methods. It describes several risk measures and risk-allocation principles, emphasizing Euler allocation to define each asset’s contribution to portfolio risk.…
IREA is an automated countertrend strategy based on the idea that unusually large price moves may be followed by movement in the opposite direction. It uses an InverseReaction indicator and enters against the shock on the next bar, provided the bar size…
This document describes an automated trading system that uses a volume-weighted moving average candle indicator. It generates a trade signal when a completed bar changes color from green to pink or from pink to green. The document does not specify which…
This document explains how an MQL5 Expert Advisor can export its trade history after a Strategy Tester run. It describes creating a history-export object, calling its export method from the tester callback, and optionally attaching the expert’s name,…
This short forum post gives a data access pattern for retrieving historical benchmark or stock data from a trade module. The example requests closing prices and volume for a benchmark symbol over a specified lookback, using daily frequency, and assigns the…
The indicator estimates volatility using ATR, a Parkinson high-low estimator, or close-to-close return variation, then ranks the current reading against a rolling history as a percentile. Thresholds divide that percentile into five regimes, from unusually…
This engineering guide explains how to build Rust-native adapters that connect NautilusTrader to exchanges and data providers. It covers venue-specific data and execution clients, configuration and Python exposure through PyO3, plus contracts for…
ResSup is a chart indicator that plots two lines derived from local price extremes. Its described trading rule is to buy when price crosses above the upper line and sell when price falls through the lower line. The lookback period controls how many bars are…
The report describes a stock-selection strategy that predicts the future usefulness of seven style factors and adjusts their portfolio weights over time. It uses historical factor information coefficients (ICs), macroeconomic variables, and market variables…