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.
21,023 documents
This post describes a stock screen requiring a daily price range greater than 1%, a trading date in 2021, and float market capitalization above 10 billion yuan. It presents these conditions as a way to identify higher-volatility shares while limiting the…
This note presents a third-generation moving average attributed to Dr. Manfred G. Dürschner. It calculates a short-period average, then averages that result over a longer period, and combines the two with a coefficient derived from their relative lengths.…
This report evaluates the Shanghai 180 index and the Hua An Shanghai 180 ETF. It describes an index weighted heavily toward financial companies and food and beverage firms, with large, liquid constituents. The report characterizes its recent one-, three-,…
The Absolute Strength Oscillator (ASO) is described as a trend-following indicator that compares bullish and bearish forces to represent the current market trend. Its configurable inputs include the calculation period, smoothing period, mode, calculation…
This article surveys three uses of unsupervised learning in investment research: manifold learning, clustering, and matrix factorization. For visualization, it applies t-SNE to fund returns, projecting high-dimensional observations into two dimensions so…
The document describes a script intended to compare how much an asset moves across trading days. Its single input is the number of trading days to include. The proposed workflow is to analyze historical data and convert the output into a column chart, making…
This research summary tests whether hedge funds can time sector-specific returns and whether that ability predicts subsequent performance, flows, and survival. The study first removes market exposure from industry returns, then estimates fund-level timing…
The essay argues that a strategy’s statistical edge depends on applying its rules consistently. It distinguishes knowledge of a method’s behavior and risks from knowledge of one’s own ability to follow it under pressure. A hypothetical stock entry missed…
The document describes a smooth-step indicator as a way to map values into a normalized range from zero to one. It places the function among sigmoidal or clamping transforms and relates its behavior to the stochastic oscillator: the indicator produces a…
This notebook demonstrates an Alphalens workflow for evaluating a daily stock factor based on the gap between the prior close and current open. It defines an example universe of large-cap equities with sector labels, calculates the gap, and aligns the factor…
This Chinese equities screen narrows its universe to stocks classified in the metaverse sector, then looks for positive returns and high ranking by large-order net amount. The article presents the flow ranking as a way to identify potential buying interest.…
This A-share stock screen combines daily price movement, relative trading activity, listing age, and the opening price. It selects stocks whose intraday high-to-low range is at least 1% of the low, whose volume is between 1.5 and 6 times its five-day…
The document explains a rolling estimate of annualized variance using daily stock returns. First compute each daily return from consecutive closing prices, then calculate the variance over a 20-trading-day window and scale it by 250, an assumed count of…
This article summarizes a study of machine-learning methods for fundamental equity valuation across 17 European countries. It estimates monthly fair values from 21 accounting variables, then defines a mispricing signal as the gap between model-implied value…
This research summary compares linear valuation models with machine-learning methods for estimating the monthly fundamental value of stocks in 17 European countries. It constructs a mispricing signal from the difference between estimated fair value and…
The article presents five claimed strengths of quantitative investing: building models from processed data and backtests, using machine learning to handle information, applying systematic rules to reduce emotional decisions, analyzing broad datasets, and…
This article explains why individual investors may find it difficult to compete with quantitative trading firms. It highlights four claimed advantages: automated rules can reduce emotional decision-making; computational systems can monitor many securities…
The document proposes a Chinese equity screen that combines a prior-day appearance on the market's notable-trading list with a price-range filter and ownership-concentration rankings. Its stated criteria are daily amplitude above 1%, a prior-day notable-list…
This research note explains how genetic programming can evolve mathematical formulas from market data to discover equity selection factors. Starting with randomly generated expressions, the algorithm evaluates their fitness against a target and applies…
This document explains how to trade a cointegrated pair using entry and exit levels produced by a minimum-profit optimization method. It defines a spread from the two asset prices and a hedge coefficient. When the spread falls below the buy threshold, the…
The document explains a regression trend indicator that gives less influence to bars with high local residual variance. It first fits an ordinary least squares line, smooths squared residuals to estimate local variance, adds regularization, and converts…
This Chinese A-share stock screen combines a technical condition with fundamental growth and liquidity filters. It selects stocks with RSI below 65, parent-company net profit growth above 20% and no more than 100%, and prior-day trading value above 60…
This paper compares machine learning methods for forecasting equity returns across the market time series and the cross section of stocks. It frames risk premium measurement as a prediction problem and describes how high-dimensional predictors,…
This Chinese A-share screening rule selects stocks with turnover between 3% and 12%, a positive product of the day’s price change and net large-order volume, and prior-day trading value above 60 million. The document presents these conditions as a way to…