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 brief description outlines a support and resistance indicator built from rolling minimum and maximum prices over a chosen number of bars. It uses those extrema to form average levels, then combines multiple levels to suggest the prevailing trend and…
This research examines whether large and small order money flows predict cross-sectional stock returns. It standardizes net inflow by trading value to form separate factors, then compares stocks grouped by factor strength. The reported results show a…
This introductory overview explains neural networks through the progression from a single perceptron to deeper architectures. It describes weighted inputs, activation functions, loss minimization, and gradient descent, noting how learning rate affects…
This short note raises a basic question about what a probability estimated by logistic regression represents. It contrasts interpreting a prediction as the chance that one particular person succeeds with interpreting it as the expected success rate among…
The study tests whether hedge funds can time returns in particular industries and whether measured timing ability predicts later fund outcomes. It first removes market exposure from industry returns, then estimates each fund’s industry timing coefficient…
This Backtrader example demonstrates how to load minute-level CSV data for a chosen date range and trading session, optionally filter records to session hours, and fill missing bars within that session. The filler can assign a configurable volume to…
This note explains the idea behind John Ehlers’ Instantaneous Trend Line. It argues that a simple moving average spanning the dominant cycle can cancel much of that cycle: samples above and below the midpoint balance across a full cycle. In frequency terms,…
This brief reference describes a tick-based stochastic indicator as a modified form of the bar-based stochastic. Its adjustable settings include the K and D periods, smoothing, calculation methods for both lines, applied price, and overbought and oversold…
This document summarizes a method for evaluating mutual fund managers while accounting for both factor exposures and the characteristics of the stocks held. Its premise is that standard factor models, including the Carhart four-factor model, may leave return…
This tutorial explains how to assess strategy performance by examining completed round-trip trades: positions opened and later wholly or partly closed. It argues that trade-level frequency, duration, and profitability can reveal whether results came from…
This overview explains feature engineering as the work of turning raw market and company data into useful inputs for quantitative strategies and machine-learning models. It uses a dual moving-average crossover to show that a trading rule can be understood…
The document describes a chart indicator that displays the ratio between a selected instrument and a benchmark instrument in a separate window. Its inputs let the user choose a calculation period, calculation method, applied price, and comparison instrument.…
This post proposes screening Chinese stocks associated with the metaverse and beverage-related industries, with a volume ratio above 1.5 and below 6. Its explanation frames the sector selection as a way to focus on market themes and the volume condition as a…
This script counts consecutive runs of bullish and bearish bars over a user-selected window and plots the distribution. Each run is represented by a signed value: positive for bullish sequences and negative for bearish ones, with the magnitude equal to the…
The document describes a rolling ordinary least squares function that takes a requested result type, an input series, a dependent variable, and a window length. It says the function can return residuals, an intercept, or a coefficient from the rolling…
The document introduces the challenge of forecasting intraday volatility across financial assets. It notes that researchers and practitioners have developed many approaches, including intraday estimators, GARCH models, and continuous-time models. Volatility…
This article explains why parameter optimization should favor a broad, stable region of good performance over a narrow peak. A broad region suggests the model can tolerate small parameter changes; a sharp isolated peak may reflect a fit to quirks in…
This tutorial introduces Seaborn, a visualization library built on Matplotlib, using Chinese stock financial statement data. It demonstrates distribution and rug plots for return on equity, joint and pair plots for examining relationships among financial…
The document outlines a process for forecasting dividends for index constituents, aggregating those estimates into an index-level total, and predicting when the dividends will be paid. Historical distributions are described as concentrated mainly in May…
MaDev is an indicator concept for locating the greatest deviation between price and a moving average. It divides observations into frames and searches within each frame for the maximum deviation, marking that point with a signal pointer. The description…
This implementation describes a bivariate Student-t copula for modeling dependence between two variables represented by uniform pseudo-observations. It explains sampling from a correlated Student-t distribution, evaluating copula density and cumulative…
The document describes a tick-chart display tool that can show tick data for price, spread, or volume. It also states that tick data can be saved to a file and later opened in a spreadsheet or another compatible program. This makes the tool relevant for…
This short MetaTrader programming note shows how an expert advisor can detect the arrival of a new chart bar by comparing the current bar count with the count saved during initialization. It stores the baseline in an integer, refreshes a second count on each…
This brief discussion asks whether a strategy’s return calculation horizon should match its holding period. The response says the answer depends on the trading logic, but that the two are generally kept consistent so a backtest represents the intended trade.…