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.
5,018 documents
These reading notes survey quantitative investing methods across stock selection, market timing, futures and options arbitrage, statistical arbitrage, algorithmic execution, and asset allocation. They outline factor selection, style and sector rotation,…
The document summarizes research that forecasts stock industry returns using lagged returns from a broad set of industries. Because including all industry predictors creates a high-dimensional regression with overfitting risk, the method uses LASSO to select…
This document describes a market direction indicator that combines six technical inputs: RSI, CCI, directional movement, Aroon, moving averages, and Supertrend. Each input is converted to a bullish or bearish state, then weighted and passed through a…
This meetup summary covers factor research, equity strategy design, AI models, and live signal timing. For factor processing, it recommends checking both the source data and the calculation logic, including data coverage and accuracy. It suggests studying…
This tutorial explains how to construct target labels for supervised learning in equity selection. It distinguishes classification labels, which group observations into discrete outcomes, from continuous regression targets. Label definitions should reflect…
The document describes how the BigAlpha 2026 final evaluates projects across factor discovery, end-to-end large models, and AI open innovation. A team's result combines a leaderboard-based component worth 30 points with a live judging component worth 70. The…
This note offers practical guidance for developing and operating AI-driven equity strategies. It recommends judging possible strategy failure by comparing live maximum drawdown with the largest drawdown observed in a long backtest that spans a full bull and…
This tutorial walks through a machine learning workflow for a virtual stock trend prediction competition using BigQuant. It covers importing competition data, converting CSV files to HDF, inspecting data, selecting input features, extracting derived…
The discussion asks how to combine a traditional stock-selection strategy with a machine-learning ranker. The user has a screening method that produces too many candidates and wants to rank those stocks afterward. The post contrasts traditional strategies,…
This overview discusses how machine learning can address high-dimensional and nonlinear return-prediction problems that strain conventional econometric methods. It frames the challenge through empirical asset pricing, including factor models, a proliferation…
The screen selects stocks using three filters: price amplitude above one, circulating market capitalization above 10 billion yuan, and prior-day turnover above 8%. The accompanying explanation frames these as filters for price movement, company size, and…
This tutorial outlines a TensorFlow LSTM workflow for forecasting stock or index returns. It recommends transforming closing prices into returns, then arranging recent observations into sequential input groups with the next observations as prediction labels.…
This research summary compares ordinary cross-validation with time-series cross-validation for model tuning on sequential data. Randomly assigning observations to folds can train a model on future observations and validate it on the past, violating the…
This reference guide describes Lumibot as a Python framework for creating, executing, and backtesting trading strategies across several asset classes and broker or data services. It outlines the strategy lifecycle, including initialization and a recurring…
The article outlines a real-time ETH/USDT trading framework that combines LightGBM for tabular inputs with a Transformer for sequential data. It describes WebSocket market-data ingestion, minute-bar construction, technical, statistical, volume, and…
This overview surveys unsupervised learning through two groups of methods: dimensionality reduction and clustering. For dimensionality reduction, it describes PCA as a way to project high-dimensional features into fewer dimensions using covariance-matrix…
This study compares conventional return labels with alternative labels based on information ratio and Calmar ratio for AI stock selection models. It argues that labels incorporating returns, volatility, and drawdowns across a period may convey more…
This note summarizes the AlexNet deep convolutional neural network described in the ImageNet classification paper. It outlines the model’s use of multiple convolutional channels, five convolutional stages and three fully connected layers, and explains how…
This strategy applies a stock-ranking model to a filtered pool of Chinese A-shares, aiming to find stocks that may rebound from depressed levels and hold up better when the broader market weakens. The target universe is the small and medium-sized board. Its…
The document discusses why high-frequency equity factors may appear strong in isolation but fail to improve stock-selection portfolios. Even after orthogonalization against common equity factors, a signal can have a high information coefficient and strong…
The document describes a problem in equity factor models: high-frequency factors can show strong overall information coefficients and long-short returns after being orthogonalized against common factors, yet fail to improve a portfolio when used for stock…
The document introduces AI driven crypto trading bots and describes their intended role: analyze market data, identify patterns, and adjust trading decisions as conditions change. It contrasts these systems with discretionary trading, emphasizing continuous…
This overview introduces VeighNa’s alpha research framework as a linked workflow for equity research: organize local market and constituent data, build factor and label datasets, train prediction models, generate scores, and test portfolio strategies. It…
This practitioner account explains how to translate a stock ranking model into daily trade lists and manual orders. The model ranks the stock universe, buys higher-ranked names, and prioritizes lower-ranked holdings for sale. The author describes how the…