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 research summary considers when event-driven stock signals should be represented as factors within a broader multi-factor model. It cites analyst recommendations, institutional research visits, and restricted-share unlocks as events that may contain…
This brief Chinese-language forum entry asks whether stock-selection backtests using random forest classification and a StockRanker ranking model can be trusted when features were not standardized or treated for outliers. The accompanying tag states that…
The document explains practical definitions for three functions used in the Guotai Junan Alpha191 factor set. Count is illustrated as summing a binary condition over a rolling window, such as counting days when the current close exceeds the previous close.…
This BigQuant support note explains why some stocks may disappear during a feature-extraction workflow. It gives two configuration suggestions: leave the feature module’s start and end dates blank when those dates should be inherited from the upstream…
The document explains what LumiBot’s HTML backtest tear sheet and companion machine-readable metrics file contain. It lists return and risk measures such as annualized and total return, Sharpe and Sortino ratios, return over maximum drawdown, maximum…
This brief indicator description presents two ways to express price movement. The first estimates the angle of a trend line against the horizontal, using minute bars along the time axis and price points along the vertical axis. The second reports the average…
The Smooth Price Oscillator (SPO) compares two recursively smoothed price series: a slower line and a faster line. Their difference forms the oscillator. The described ProRealTime implementation scales that difference using the largest recent standard…
This study examines whether supplier and customer links help explain dependence between listed companies’ returns. It constructs a directed company network from Bloomberg relationship data, resolving cases with links in both directions by choosing the…
This document describes a plotting workflow for examining Betfair backtest logs that record bid and ask volumes by runner. It extracts periodic batch volumes and cumulative imbalance, then creates three visual views: imbalance over successive updates, the…
This indicator uses the number of transactions per minute as a proxy for market participation and compares recent activity with a long moving average. Activity above the average is displayed as a high-activity phase, while below-average activity marks a…
This note outlines a stock screen combining three conditions: RSI below 65, revenue in 2021 more than 1.1 times revenue in 2018, and shrinking negative MACD bars on a 15-minute interval. It presents the approach as a mix of fundamental growth and technical…
MATonySignal is a technical indicator designed to show both trend direction and trend strength using momentum calculated from an averaged price. It colors rising trends blue and falling trends red. Bar brightness distinguishes stronger trends from weaker…
This stock-selection proposal combines a positive but limited 10-day gain, a concentration measure below 20%, and at least five overlapping moving averages. It defines concentration as the difference between a short moving average and a longer one, divided…
Linear Sinus FT is described as a currency-quote indicator that approximates sinusoidal waves and visualizes a Fourier series. The document provides only this high-level description; it does not explain the calculation, parameter choices, signal…
This forum reply addresses a stock-selection condition that compares a five-period exponential moving average with a 169-period average. It explains that the longer average requires a data history spanning at least 169 days before the condition can produce…
This report studies whether DeFi lending activity in USDC, USDT, and DAI is associated with Ethereum volatility. It estimates daily volatility with the Garman-Klass method, which uses open, high, low, and close prices, then applies Augmented Dickey-Fuller…
The document explains merge sort, a comparison-based sorting method that repeatedly combines smaller sorted groups into larger ones. It describes the algorithm as stable, meaning equal items retain their relative order, and gives best, average, and worst…
Threshold autoregression (TAR) extends a standard unit-root test to allow a series to adjust differently depending on whether it is above or below a threshold. The document illustrates the idea with the gasoline crack spread, defined as unleaded gasoline…
This research summary examines whether outside demands on independent directors weaken their oversight and advice. It identifies personal events and pressures at other companies, such as illness, leadership changes, major transactions, investigations, or…
These release notes describe additions to pyfolio, a toolkit for evaluating trading portfolios. New analyses include performance attribution to common factors, factor and sector risk exposures, rolling volatility, capacity, bootstrap uncertainty in…
The Echo Forecast indicator searches recent price history for a similar earlier pattern, then uses the price changes that followed that historical match to project a possible path forward. It compares a reference window against candidate windows in a…
This brief description outlines an expert advisor that trades at the minimum lot size and alternates between closing its existing position and opening a new one at a configurable interval. A random number determines whether the new position is long or short,…
The article examines how benchmark choice affects estimates of active mutual fund skill. It distinguishes investable index benchmarks from factor-based benchmarks, arguing that omitted common risks can bias factor-model alpha, while choosing the wrong index…
The article reviews how benchmark choice affects estimates of active mutual fund performance. It argues that factor benchmarks and investable index benchmarks are not necessarily equivalent: omitted common risks can make factor-model alpha misleading, while…