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.
3,481 documents
This meetup summary covers factor research, equity strategy design, AI models, and live signal timing. For factor processing, it recommends checking both the source data and the calculation logic, including data coverage and accuracy. It suggests studying…
The document raises practical execution questions from a live Chinese stock strategy that submits market buys shortly after the open using a VWAP setting. The trader reports fills above the observed opening price and asks whether VWAP means immediate…
The post lists proposed translations of features from one version of a quantitative research platform to another. Examples include cross-sectional percentile ranks of current and lagged returns, industry-group ranking, candle direction, rolling price and…
This tutorial explains how to construct target labels for supervised learning in equity selection. It distinguishes classification labels, which group observations into discrete outcomes, from continuous regression targets. Label definitions should reflect…
The document describes how the BigAlpha 2026 final evaluates projects across factor discovery, end-to-end large models, and AI open innovation. A team's result combines a leaderboard-based component worth 30 points with a live judging component worth 70. The…
This note offers practical guidance for developing and operating AI-driven equity strategies. It recommends judging possible strategy failure by comparing live maximum drawdown with the largest drawdown observed in a long backtest that spans a full bull and…
This monthly review assesses equity factors using information coefficients and market-neutral long-short returns. Its February 2018 discussion reports stronger results for profitability and growth measures, including return on assets, return on equity,…
This report examines price momentum in Chinese equities and organizes possible improvements into a two by two framework: refining the price or return used to measure momentum, and incorporating the path of prices or returns during the formation period. It…
This research summary examines factor crowding: the possibility that heavy investment in a factor weakens its returns or makes them less stable. It contrasts crowding measures based on holdings, which are more directly interpretable but delayed and…
The document summarizes research into whether changes in views of finance-related Wikipedia pages can provide clues about investor information gathering before trading decisions. Its central idea is that online attention may reflect investors’ early-stage…
This report outlines a weekly framework for tracking profitability and business momentum across China’s CITIC industry groups. It compares trailing-twelve-month return on equity (ROE) with its historical percentiles over one- and ten-year windows, then…
This tutorial walks through a machine learning workflow for a virtual stock trend prediction competition using BigQuant. It covers importing competition data, converting CSV files to HDF, inspecting data, selecting input features, extracting derived…
The document addresses whether a newer platform factor, netflow_amount_large, exactly replaces the legacy mf_net_amount_xl factor. The answer is that the two are not identical: the old and new money flow factors use different data sources and different…
The post raises a research question about obtaining daily exchange-rate data for several countries, including China, the United States, and Japan. Its author argues that broad equity-market timing should consider capital flows, distinguishing domestic flows…
This report summary describes a method for improving return on equity as a stock-selection factor. Starting from quarterly ROE, it derives four features from its time series: year-over-year earnings comparison, stability, persistence, and growth. The…
The discussion asks how to combine a traditional stock-selection strategy with a machine-learning ranker. The user has a screening method that produces too many candidates and wants to rank those stocks afterward. The post contrasts traditional strategies,…
This overview discusses how machine learning can address high-dimensional and nonlinear return-prediction problems that strain conventional econometric methods. It frames the challenge through empirical asset pricing, including factor models, a proliferation…
The document outlines a two-stage quantitative investing workflow: combine several signals into a single factor score, then optimize portfolio weights against a chosen objective. It names linear combination as one approach to factor synthesis and mentions an…
This report examines whether analyst expectations data can help rank industries for rotation. It groups signals into analyst coverage and forecast disagreement, the direction of earnings revisions, and the size of forecast changes. Coverage and disagreement…
This introductory guide describes quantitative trading as building rules from data and statistical analysis to generate systematic entry and exit signals. It explains positive expectancy as a favorable average outcome across repeated trades, while…
The strategy sells at-the-money call and put options on the CSI 500 index to collect time decay, opening once near the start of the session and closing all positions before the close. It filters for contracts more than ten days from expiry and adequate…
The article introduces time series concepts for financial analysis, including univariate and multivariate data, stationarity, autocorrelation, trend, seasonality, and decomposition. It describes using ACF and PACF to examine lag dependence, and outlines a…
The document outlines ways to set the changing asset allocation, or glide path, of a target-date fund. One approach uses an investor’s human capital and forecasts of future asset returns to choose a path that maximizes investor utility. Another bases the…
This brief Q&A considers why correlations between peer companies can look different when calculated from daily returns rather than price levels. The question arises in an effort to identify the closest competitors to a target company using stock data from…