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 webinar overview introduces several Bitcoin and Ethereum measures that researchers might turn into trading signals. It groups them into market-cycle or valuation indicators, including Pi Cycle, Realized Cap, and the Bitcoin Yardstick; sentiment measures…
This analysis examines Bitcoin’s role in a diversified portfolio before and after the market’s financialization, using 2013–2017 and 2018–2023 as separate periods. It compares historical returns, volatility, drawdowns, and correlations, then applies…
This document outlines a stock screen using four conditions: amplitude above one, current volume above ten thousand lots, an opening price above the prior close, and revenue in 2021 more than 1.1 times revenue in 2018. It presents the combination as a way to…
This research summary reviews stock-selection factors derived from tick-by-tick trade data, including large-buy share, buy-order concentration, intraday aggressive buying, and informed buying or selling measures. After orthogonalization, the factors…
This document describes a modified Phase Change Index (PCI) that adds Jurik Moving Average smoothing. It presents three configurations: smooth the price data before calculating PCI, smooth the calculated PCI, or apply smoothing at both stages. The change is…
The document outlines a classic radial basis function neural network with a hidden layer of radially symmetric neurons and a linear or sigmoid-style output layer. It describes choosing the output activation from the training target range: hyperbolic tangent…
The document describes a study of four MSCI smart beta equity indexes and a market weighted benchmark. It uses principal component analysis (PCA) to separate shared market movements from strategy specific components, then regresses those components on Fama…
This document is a QuantStats tear sheet comparing an AI trading strategy with SPY over a brief January 2026 backtest, using Yahoo data. It reports a 1% total return for each, with the strategy showing a higher annualized return estimate but also a larger…
The post describes a stock filter using three conditions: turnover between 3% and 12%, external trading volume divided by internal trading volume above 1.3, and a share price above 18.5. It treats turnover as a measure of trading activity and the volume…
The document is a country-level table of oil production economics. It lists operational cost, capital cost, total cost, estimated reserves, daily production, and the 2015 average oil price for twenty producers. The figures allow readers to compare reported…
This indicator description explains a tool that plots Repulse values in a separate pane, compares the oscillator with price, and marks divergences. Users can choose whether divergence detection follows the main Repulse oscillator line or its signal line. The…
This Python module provides diagnostics and configuration helpers for a systematic trading system. It compares each rule’s capped forecasts and each instrument’s combined forecasts with a target average forecast magnitude, ranking the largest discrepancies…
This Chinese equity screen selects stocks with turnover between 3% and 12%, year-over-year net profit growth attributable to parent-company shareholders above 20% and up to 100%, and revenue in 2021 more than 1.1 times its 2018 level. The document supplies…
This research summary reviews a study of how unexpected Federal Reserve policy changes relate to returns across firms with different financing constraints. Using U.S. company and stock data from 1994 to 2007, the study classifies firms by constraint indices…
This Chinese-language post collects questions about factor research and strategy implementation, but does not provide answers. Topics include adding data factors, scheduling entries and exits, evaluating and combining factors, handling collinearity,…
The document introduces a Nadaraya–Watson kernel regression indicator that applies Gaussian weighting to historical prices. It describes how the method gives more influence to nearby price observations and uses the resulting smoothed estimate to form a…
This equity screen selects main-board stocks with turnover between the stated bounds, circulating market value within a specified range, and more than a year since listing. The document presents turnover as a liquidity-related filter and float value and…
This document proposes selecting stocks whose daily price amplitude exceeds 1%, whose MACD is above zero, and whose previous-day actual turnover falls between 3% and 28%. It describes amplitude as a measure of price movement, MACD as a trend reference, and…
The article introduces the Fama-French five-factor model as an extension of the three-factor framework. It describes market, size, book-to-market, profitability, and investment factors, and explains that multiple regression estimates a stock’s exposure to…
This note presents a relative strength index calculation based on closing-price changes. It separates each change into its positive component and its absolute magnitude, then applies a smoothed moving average to each series. The ratio is transformed onto a…
The document reviews quality investing as a composite of company characteristics, including profitability, growth, stability, capital structure, accounting quality, and investment. It compares reported findings for U.S. equities with cross-sectional results…
This short community thread raises possible future-data problems in an equity screening and backtesting tool. One commenter warns that sorting by current market capitalization may select companies based on their later size, making historical results appear…
The document surveys data used to build quantitative investment strategies and support trading and risk decisions. It groups inputs into market prices, volumes and trades; company financials and valuation; macroeconomic indicators and policy; technical…
This brief indicator description presents a way to visualize price series processed by a Kalman filter as a sequence of candlesticks. The filtered series is rendered in candle form, which the source suggests may make the resulting analysis more informative…