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 document explains how time-series momentum (TSMOM) and moving-average crossover (MACROSS) strategies can represent closely related ways to measure price trends. It describes expressing trend signals through past prices or returns, and using “signature”…
This document describes a Chinese A-share selection screen combining robot-industry exposure, small free-float market capitalization, turnover, and the stock’s price change at 9:25. The stated selection logic calls for turnover between 3% and 12%, robot…
A platform forum exchange addresses whether a daily backtest that predicts a five-day return using the current day’s closing price can access that day’s prediction before the close is available. The response explains that, in the described daily setup, the…
This guide presents local checks for two common failures in factor submissions: missing values near the evaluation window’s start and look-ahead bias. For time-series factors, it recommends querying enough earlier data to cover the longest lag or rolling…
The article summarizes research that uses active share, or portfolio deviation from a benchmark, as a proxy for a fund manager’s conviction. It reports that strong past performance is associated with subsequent increases in active share, particularly for…
This dated market note reports RSRS timing-model views for several Chinese equity benchmarks on April 8, 2020. The model remained bullish on the Shanghai Composite, CSI 500, and ChiNext, while taking a more cautious view of the SSE 50 and CSI 300. No…
The article presents regular investing in broad market index funds as a low-effort, long-term approach, arguing that it can reduce the need for stock picking and market timing. It attributes active managers’ difficulty beating indexes to fees, short-term…
The document summarizes research on combining market anomalies and on refining multi-factor models. For anomaly strategies, it reports that returns to many well-known anomalies declined after academic publication, while diversified long-short portfolios…
This page summarizes several quantitative investing topics: timing equity factors with cross-sectional and time-series indicators; adjusting portfolios to target volatility; modeling changing stock-bond correlations; assessing tactical asset-allocation…
This note summarizes when to use several portfolio allocation approaches. A naïve allocation is presented as a practical choice when the universe is broad and return forecasts are difficult. Risk parity and minimum variance are associated with settings where…
This Chinese-language summary presents a 2021 survey of reinforcement learning in financial decision-making. It introduces Markov decision processes as a foundation, then identifies value-based and policy-based learning approaches, including extensions that…
This strategy combines model-based stock ranking with technical timing. It starts from a manually defined CSI 150-related universe, then ranks candidates using an AI prediction score and volume, turnover, and large-order money-flow factors. Timing filters…
This guide surveys how machine learning can support quantitative investing, from data preparation and model evaluation to portfolio selection. It explains supervised, unsupervised, and reinforcement learning, and reviews methods including regularized…
This Chinese strategy discussion combines hand-built stock selection rules with a CNN classifier. It describes converting screening conditions into Boolean features so a model can learn to identify stocks expected to show short-term strength. The feature…
The document presents a short-term timing approach based on the Hilbert transform. It assumes price movements contain a long-term trend, shorter cyclical fluctuations, and noise. The proposed process first smooths the data with a 20-period moving average,…
This weekly Chinese equity factor note compares value and growth styles and reports relative performance across six industry-neutral long-short factor groups. Value and growth were described as broadly balanced, while quality measures such as return on…
The article describes combining two Chinese equity selection models designed for different market conditions: a rebound after an index decline and a mild index pullback. Each model includes market environment features so that it produces candidates only when…
This research summary describes a dynamic rebalancing overlay for multi-factor equity portfolios. The portfolio keeps its regular monthly rebalance, while market-cap factor exposure is checked daily mid-month. If exposure breaches a set threshold, the…
The post contains a proposed Chinese stock screen built from financial, valuation, price-action, and intraday-style indicators. Its filters include market-cap and asset thresholds, revenue growth and return on equity conditions, turnover and price-change…
The article explains genetic algorithms as population-based search methods inspired by biological evolution. It walks through encoding candidate solutions as chromosomes, evaluating them with a fitness function, selecting parents, combining them through…
This study applies Lasso, adaptive Lasso, and Elastic Net to select equity factors and combine them into stock-selection signals. Its motivation is that correlated predictors can make ordinary multivariate regression unstable. The reported experiments show…
The post asks whether a factor-analysis module’s rebalance period measures returns before or after the factor observation. The author expected each factor-sorted group’s reported return to represent the return earned by holding its stocks for the stated…
This April 2020 brokerage-report summary argues that overseas pandemic concerns had weakened prices without materially lowering broad-market valuations or company earnings forecasts at that point. It frames semiconductor shares against a late-2018 price…
This document summarizes Ibbotson and Chen’s supply-side approach to estimating long-run equity returns and risk premiums. It decomposes US stock returns over 1926–2000 using inflation, earnings, dividends, valuation multiples, payout ratios, book value,…