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 reference lists configuration options for FreqAI, Freqtrade's machine-learning feature. General settings cover rolling training and inference windows, model identification and persistence, retraining frequency, model expiration, and prediction…
This tutorial explains how to aggregate one-minute bars into longer intraday bars, using a Chinese stock session as its example. It shows the standard time-series resampling approach: assign open from the first observation, high and low from the extrema, and…
The article explains support vector machines (SVMs) as classifiers that seek a separating boundary between classes. It introduces the geometric intuition of mapping observations into a higher-dimensional feature space, where a nonlinear boundary in the…
The report examines how changes in market interest rates can affect listed companies’ net profits through interest expense. It compares industries by the importance of financial expenses relative to net profit, finding higher exposure in areas such as light…
This brief forum exchange addresses a quant workflow problem: a ZScoreNorm standardization step produces only missing values. The response explains that the operation performs cross-sectional normalization and suggests inspecting the input data for an…
This summary of a Chinese brokerage factor-monitoring report compares recent performance across equity factors using several views: long-only excess returns, long-short returns, information ratios, and relative strength over roughly six months. Factors…
The article argues that durable trading depends on treating each trade as a planned decision rather than an impulse. It recommends defining entry rules, recording the reason and timing for each trade, waiting for qualifying signals, and setting exit plans…
This forum post raises a factor-data debugging question: why can an industry-neutralized price-to-book factor contain missing values when the raw factor appears complete? It identifies an industry-neutralization function applied using a first-level industry…
The indicator estimates current and upcoming daily volume by averaging historical observations from the same weekday. Users choose the number of weeks to include and whether to display tick volume or real volume. The plotted average can be compared with…
This article argues that LSTM models trained on historical prices can appear to forecast stocks accurately because predicted daily prices often remain close to the previous day’s price. When plotted against actual prices, this persistence can look…
This document describes an MQL5 class for calculating Wilder’s Average Directional Movement Index with a ring buffer. It explains the initialization settings, including the smoothing period and method, buffer capacity, and input indexing mode. The class…
The article explains backtesting as a way to replay trading rules on historical data, check whether signals and logic behave as intended, and expose strategy defects before deployment. It emphasizes that historical simulation can test market assumptions, but…
This forum post reports a data-quality anomaly in the CSI 1000 constituent history: a query finds 1,001 component records for the index on July 26, 2023. It shows how to query the component table for dates from 2015 onward, then group records by date for the…
This summary of a research report examines whether trade counts from tick-level transaction data can refine a traditional reversal factor in China’s order-driven A-share market. Trade count is defined as the number of matched transactions. The report says it…
The article summarizes a method for extracting priced common factors from stock returns. First, a Fama-MacBeth cross-sectional regression uses several return-predictive characteristics to estimate stocks’ expected returns. Stocks are then sorted into…
This article outlines a reproduction of a research approach that predicts whether a stock’s closing price will rise or fall over a future horizon. It applies exponential smoothing to Apple daily price data, then builds a binary target from the price change…
This report excerpt describes a weekly framework for monitoring profitability and business conditions across selected primary and secondary industries in China's CITIC classification. It tracks trailing-twelve-month return on equity and places current…
This tutorial introduces bias, variance, and irreducible error as sources of prediction error. It relates model complexity to underfitting and overfitting: overly simple models tend to perform poorly on both training and test data, while overly complex…
This document describes a bootstrap test for judging whether a strategy’s average closed-trade result appears distinguishable from chance. It repeatedly resamples the observed trades with replacement, computes the average for each resample, and uses the…
This brief programming note explains how to convert numeric timestamps in JSON into calendar dates. It uses sample values that pair large timestamp integers with compact date representations, then shows the general conversion approach in JavaScript and…
The document introduces cointegration as a long-run relationship in which individually nonstationary series can have a linear combination that is stationary. It explains stationarity through stable statistical properties over time and contrasts it with…
The article argues that mean and variance alone may miss information when market returns depart from a stable Gaussian pattern. It describes higher moments, especially odd-order moments, as potentially useful timing signals: their values reportedly expand…
This proposed Chinese equity screen combines a price movement threshold with the presence of an outstanding convertible bond name and positive prior-session auction turnover. The prose describes auction turnover above 0.26, while the formula expresses the…
This discussion asks whether the maturation of China’s equity market will make quantitative investing harder. The concern is that institutional participation and competition may reduce behavioral inefficiencies and compress excess returns for both systematic…