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 note offers informal Sharpe-ratio thresholds for screening strategies under a stated evaluation context: a single strategy rotating half-capital among at least three stocks, with million-level capacity and a backtest covering roughly three years. It…
This document describes a parameter sweep for a grid trading backtest. It combines every configured symbol with candidate relative half-spread and grid-count values, then runs the resulting backtests in parallel over a selected date range. The grid interval…
The document describes adapting an automated trading system to test signals from a ZigZag-based trend indicator. The system opens trades when signals arise and exits through stop-loss or take-profit orders; it can hold multiple orders, with the count tied to…
The strategy ranks a broad set of commodity futures by their performance over the previous 12 months, divides the available contracts into quintiles, and holds long positions in the strongest quintile and short positions in the weakest. It rebalances monthly…
This overview explains the main components of a hosted quantitative trading platform: a control center, strategy library, public live-trading pages, simulated trading environment, community, and API documentation. It describes how users can store strategies,…
The document explains the behavioral challenges of trend following: traders must tolerate repeated small losses, fluctuations in open profits, and periods of waiting for a large move. It recommends evaluating a strategy over many trades rather than judging…
This document presents a Chinese stock screening rule that combines a five-period amplitude threshold, a change in a proxy for institutional trading activity, and a newly formed bullish KDJ crossover. The crossover condition requires the J line to move above…
The document introduces the Sharpe ratio as a measure of excess return per unit of portfolio risk, then describes a method for forecasting how that ratio changes over time. It outlines an approach associated with Whitelaw that estimates Sharpe ratios from…
This BigQuant example builds a binary classifier from rolling windows of daily open, high, low, close, and volume changes. Its network applies a convolutional layer to each 50-step, five-feature input, passes the result through an LSTM and dense layers with…
This Chinese equity screen filters for stocks in the beverage and alcohol import-export sector, with amplitude above 1 and a positive ten-day return below 35%. The stated rationale is to focus on shares showing price movement and recent gains that have not…
This report presents a brief backtest of a market-news trading bot against SPY, covering January 4–15, 2026. It lists return, drawdown, risk, correlation, and other performance statistics, along with model-call and data-source details. The strategy reports a…
This post presents a Chinese A-share stock screen based on daily price range and a large-order net-volume ranking, while excluding Beijing-listed stocks. It frames the approach as a way to find active stocks and potential medium- to long-term opportunities.…
This review evaluates an RSRS market timing strategy out of sample across several Chinese broad-market indices from March 2017 to November 2019. It reports generally positive timing results, strongest on the CSI 300, while the CSI 500 results were more…
This research note outlines a framework for evaluating funds and constructing allocations across active equity funds, enhanced index funds, and active bond funds. For equity funds, it discusses classification and performance attribution using both holdings…
This strategy ranks a universe of ten US sector ETFs by their trailing 12-month rate of change. At each monthly rebalance, it selects the three strongest performers, assigns them equal portfolio weights, and exits holdings that no longer qualify. The…
The document presents a Chinese A-share stock screen combining three conditions: a 14-period RSI below 65, at least two limit-up sessions within the prior 500 days, and a positive ten-day return below 35%. It describes the filter as a way to find stocks with…
The document describes a low-lag trend line (LLT) designed to reduce the delay that comes with smoothing price data. It frames moving averages as a trade-off: longer windows smooth noise more effectively but respond more slowly, while exponential averages…
This document outlines a Forex and stock trading Expert Advisor that combines Fibonacci levels with two moving averages. It describes trading around price breakouts, with Fibonacci levels serving as a filter alongside the moving averages. The EA is said to…
This document describes an Expert Advisor that trades signals from the ThreeCandles indicator. It opens a trade when a colored indicator candle appears at the close of a price bar. The EA depends on a compiled indicator file, and a supporting library is…
This reference explains how Freqtrade handles pair naming, fees, and strategy execution. Spot pairs use a base and quote currency, while futures pair names also identify the settlement currency. Profit calculations include fees: simulations use the…
The document summarizes research on time-series intraday momentum in US equity ETFs. It describes a signal in which the first half-hour return, measured from the prior close, predicts the final half-hour return in the same trading day. Using SPY data, the…
The document introduces attention and Transformer components, then adapts them to forecast stock returns from factor sequences. Its proposed architecture adds Time2Vec features to factor inputs, applies self-attention and encoder layers, and uses a…
This forum exchange explains where to look when a VeighNa strategy backtest produces no visible trade records and a developer wants to inspect daily messages. The reply says scripted backtests expose log entries through the engine’s logs collection. For live…
These notes summarize the Turtle trading approach as a systematic trend-following method built around breakouts, disciplined execution, and survival through losing periods. The described rules include entering when prices exceed a prior-period high or low,…