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 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 document explains a chart heatmap that compares an instrument’s typical volume or volatility across months, days of the month, weekdays, or hours. It calculates each period’s total volume or true range, groups observations by the selected calendar…
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 indicator description explains how a sinusoidal moving-average tool exposes phase, opposite phase, and amplitude as separate plotted values. Phase is represented in degrees across a full cycle, while the opposite phase runs through the negative range;…
This brief document describes a charting script that calculates and displays the autocorrelation function (ACF) and partial autocorrelation function (PACF) for a selected data window. Its inputs control the number of observations, the number of lags, the…
This Chinese-language post proposes screening stocks for a watchlist using three conditions: a daily high-low range exceeding one percent, at least one limit-up event within the prior 25 trading days, and a 2019 dividend payout ratio above 25 percent. The…
The document introduces a zero-lag variant of the T3 technical indicator. It describes T3 as a smoothed indicator designed to respond relatively quickly, then says the variant incorporates calculation approaches associated with Tim Tillson and with Bob Fulks…
This note describes a Chinese equity screen that ranks stocks by estimated capital strength, looks for substantial net institutional inflows, and restricts the universe to companies listed in 2021. It suggests using turnover and trading volume as proxies for…
This document describes a MetaTrader indicator that smooths a recursively calculated band-pass series to produce a trend component. Its inputs are the calculation period, a delta factor used to derive gamma, and the price series. The recurrence uses current…
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 brief indicator note discusses Detrended Synthetic Price, an indicator attributed to John Ehlers. It argues that using fixed levels to interpret the oscillator can be cumbersome because the appropriate levels may need adjustment for each instrument,…
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…
This article describes a Chinese equity screen that selects stocks with turnover between 3% and 12%, a positive product of price change and net large-order volume, and excludes Beijing-listed shares. Its sample indicator and Python material illustrate…
This article proposes a Chinese equity screen that looks for stocks with a daily high-low range above a threshold, no limit-up close on the previous day, and signs of large-investor control. It gives an indicator formula for price range and a sample Python…
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…
The document introduces nine indicators that compare historical range, volume, and equity gains or losses across calendar periods. For each measure, the collection examines months, weeks, and days of the week to identify periods that have tended to be above…
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,…
The document summarizes research on improving a primary sequence-learning task by selecting among related tasks that differ in timing. Examples include translation with different input delays and stock trend prediction over different future horizons. Because…
The document introduces linear regression as a way to describe how one variable changes in relation to another. It frames the model as a fitted straight line, with an intercept and slope, and uses daily returns of a Chinese bank stock and the CSI 300 as an…
This analysis proposes heterogeneous volatility as an alternative to the standard volatility factor in multi-factor equity selection. Standard volatility is the dispersion of a stock’s returns. The proposed measure instead captures the volatility of returns…
This indicator description explains a MACD-style histogram built from two idealized moving-average indicators, then smoothed using one of ten averaging methods. The choices range from simple, exponential, smoothed, and weighted averages to adaptive or…
ATRNorm is described as a normalized Average True Range indicator intended to identify periods of relatively flat price action. Its inputs include the lookback period, a smoothing period, the type of series to normalize, a normalization window, volume type,…
The document explains feature engineering as the process of turning raw data into variables that better represent a prediction task. It argues that useful features matter in quantitative strategy research because market data and factor choices constrain what…