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 author reflects on losses in leveraged forex trading and explains two staking approaches using a fair coin game. A martingale raises stakes after losses and depends on effectively unlimited capital; a reverse martingale risks a fixed fraction of current…
This document introduces an MQL5 port of ALGLIB, a cross-platform numerical analysis library, and outlines how its packages can support quantitative research. Covered methods include linear algebra, equation solving, interpolation, optimization, Fourier…
The document introduces decision trees for classification and regression. It explains how a tree recursively partitions observations using feature tests, with branches representing outcomes and leaves providing class predictions or numeric estimates. Tree…
This overview explains statistical arbitrage as a family of strategies that trade relative mispricing across related instruments. It distinguishes cross-market, cross-asset, ETF, and market-neutral approaches, and gives pairs trading as a central example:…
This event study examines how Chinese convertible bonds and their underlying shares behave around conversion, early redemption, and conversion-price resets. It reports that conversions increase share supply and can weigh on the underlying stock. Early…
This research digest reviews two portfolio topics. The first concerns Shiller’s cyclically adjusted price-to-earnings ratio (CAPE): it can help forecast long-run equity returns as valuations tend to mean-revert, but its short-term signal is unreliable when…
This brief platform support exchange distinguishes missing feature values from NaN values produced by calculations. The support response says missing values had already been filtered; remaining NaNs were calculation results, such as an undefined zero divided…
This document describes an automated or manual pair-trading robot that opens positions in two symbols when their correlation meets a configured threshold. It distinguishes pairs, whose charts move similarly, from mirror symbols, whose charts move in opposite…
The document explains the maximum drawdown figure shown on Hyperliquid’s portfolio page and clarifies that it is a convenience metric displayed by the frontend. It does not affect margin requirements or other platform calculations. To reproduce the displayed…
The document summarizes a study of Alpha101 factors in China’s A-share market and a method for generating additional price-volume signals with genetic programming. It says the authors assessed existing factors using Rank IC, portfolio stratification, and…
This stock screening proposal selects shares with turnover between 3% and 12%, seven consecutive days of falling prices, and positive net buying attributed to major market participants on the prior day. The rationale is to look for heavily declining stocks…
This screening proposal selects Chinese A-shares with turnover between 3% and 12%, excludes Beijing-listed shares, and requires the 9:25 indicated gain to be below 6%. It presents turnover as a way to target trading activity and liquidity while avoiding a…
This article presents an equity screen combining three conditions: price range greater than one, return on equity above 15% in each of the prior five years, and turnover between 3% and 12%. It frames the filters as a way to find companies with sustained…
This report summary explains a framework for estimating returns from IPO subscriptions in China, in the context of the first companies listing after the ChiNext registration reform. It identifies three drivers of subscription returns: the amount raised, the…
This document introduces a volatility indicator intended to distinguish quieter from more volatile market conditions. It motivates the approach by noting that standard deviations can measure variation, but fixed thresholds may not provide useful boundaries…
This Chinese-language article introduces order flow toxicity through the market microstructure problem of information asymmetry. Informed traders may profit at liquidity providers’ expense; if this adverse selection becomes severe, market makers may withdraw…
This FAQ describes the scope and practical use of an educational financial reinforcement learning library. It covers supported data sources, feature inputs such as sentiment, training options, reward functions, hyperparameter tuning, and algorithm choices.…
The document explains the correlation coefficient as a measure of how two financial instruments move together. Values near positive one indicate similar movement, values near negative one indicate opposite movement, and values around zero indicate little…
This forum post raises a data-preparation problem in an LSTM trading strategy. The author reports that filtering rows before rolling sequence windows can remove observations that should remain part of the temporal context, causing later observations to shift…
This post asks how to obtain moving averages covering seven, fifteen, and thirty days through a trading platform’s data and technical-analysis functions. It shows sample calls that request records at several candle intervals and then apply moving-average…
The article compares two ways to build equity strategies: train a predictive model on chosen factors and a defined stock universe, or specify screening conditions directly. The AI approach can be quicker to assemble, but its behavior may be hard to diagnose…
The document describes an indicator that calculates the Pearson correlation between a chart symbol and a user-selected second symbol. The user can configure the comparison period and applied price, along with minimum and maximum correlation settings that…
Fundamental quantitative investing applies statistical methods and algorithms to company accounts, industry information, and macroeconomic data to rank stocks by estimated value or quality. The document describes analyzing profitability, financial strength,…
The paper examines how five emotions expressed on Sina Weibo—anger, disgust, fear, joy, and sadness—relate to Chinese stock returns. It uses machine learning to classify posts and wavelet analysis to study how the relationship changes over time and across…