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
The report develops stock-selection factors from one-minute Chinese equity data, adding transaction counts to price and volume inputs. It examines average trade value and the direction of large-trade activity, then evaluates these signals both alone and…
This platform guide distinguishes raw financial statement records from processed point-in-time (PIT) data and explains why the distinction matters in quantitative investing. Raw records are organized by instrument, report date, publication date, and change…
This monthly review compares equity factors using information coefficients (IC) and long-short returns. It reports that growth and turnover factors were relatively consistent over the preceding three months, while profitability and reversal results varied;…
This research summary studies whether lagged returns from related industries can help predict industry returns for sector rotation. It motivates the question by noting that such predictability would not be expected in an ideal frictionless market, while real…
This report summary describes a Barra-style risk model for Hong Kong stocks, built to study factor returns and portfolio exposures. Its data foundation combines security details, prices, trading activity, Hang Seng industry classifications, valuation…
The document gives a brief definition of the harmonic mean. For a series of values, it is calculated by taking the number of observations and dividing it by the sum of their reciprocals. Equivalently, it is the reciprocal of the arithmetic mean of the…
This documentation explains how to configure FreqAI within a Freqtrade configuration and strategy. It outlines core settings for training and backtesting periods, model identification, timeframes, correlated pairs, shifted candles, indicator periods, labels,…
This forum post describes a reproducibility question in a quantitative modeling workflow. The author sets a learning-rate parameter to the same value in the model and in a hyperparameter-search module, and configures the search grid to contain only that…
The document explains sliced inverse regression (SIR) as a supervised dimension-reduction method for financial prediction. Unlike principal component analysis, which summarizes variation in predictors without using the target, SIR uses the forecast variable…
This introductory explanation presents support vector machines as supervised classification models. In a simple two-class example, many lines can separate the data, but SVM selects a boundary that maximizes the margin between the two classes. The closest…
The document describes a daily Chinese-equity stock screen combining price action, trading activity, market capitalization, and profitability. The proposed filters require an intraday range above 1% relative to the prior close, market capitalization below 10…
This Expert Advisor combines a random-number gate with a moving-average price relationship to decide whether to open a position. The random draw is compared with one of eight thresholds represented by poker-hand labels; if it falls below the selected…
This overview surveys models used to relate asset values or expected returns to risk and other inputs. It describes CAPM as linking expected return to the risk-free rate and market exposure, then introduces multifactor models such as Fama–French, which add…
QQECloud is described as an indicator built from smoothed relative strength index calculations. It is presented in two forms: a two-line QQE display and a color cloud variant. The document identifies QQE as a quantitative and qualitative evaluation method,…
The Pivot Oscillator is defined from three moving-average differences applied to typical price, calculated as the average of a bar’s high, low, and close. The method forms differences between fast, medium, and slow moving averages, adds those differences,…
This reference describes a library of fast Fourier transform routines for processing sampled signals. It covers forward and inverse transforms for complex and real-valued data, a combined transform for two real signals, discrete sine and cosine transforms,…
TD_I is presented as a modified DeMarker oscillator attributed to Thomas DeMark. It scales readings to a 0–100 range and compares the current bar’s high and low with those from a configurable earlier bar, rather than always using the immediately preceding…
The Price Momentum Oscillator (PMO) converts one-period price change into an oscillator by applying two successive custom exponential-style smoothing steps. The first operates on the percentage change; the second smooths a scaled version of that result. A…
The article explains why implied volatility is an option price input solved from a pricing model, rather than a direct forecast of future realized volatility. It recommends assessing current IV against its own history and reading it alongside the shape of…
This document records a failure in a stock-ranking training module. The run loads 6,711 training rows, begins data and feature preparation, then stops with an IndexError while calculating maximum discounted cumulative gain. The traceback points to code that…
A user questions a discrepancy between a stock’s reported historical prices and the value returned by a platform’s current-data function. For a Chinese-listed equity on one trading date, the user reports historical opening, closing, and low prices that…
The distance approach forms pairs by rescaling each asset’s training-period prices to a common range, calculating the sum of squared differences between each pair’s normalized series, and selecting the closest matches. In the cited original study, the…
This guide introduces statistical learning through an example of predicting the S&P 500 from company fundamentals. It frames the task as estimating a response from predictor features plus residual error, then distinguishes prediction, which prioritizes…
The document describes a script that writes closing prices for all symbols available on a trading platform to a CSV file. The user can choose the output filename and select which bar’s close to export, making the script a basic way to collect a cross-section…