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
The document describes a high-risk, high-turnover strategy for Chinese stocks under the T+1 trading convention. It looks for strong stocks that pull back and rebound, possible second moves in leading stocks, and sentiment-driven stocks that may reverse after…
This article explains how to model multi-stock trading as a Markov decision process and train a deep deterministic policy gradient (DDPG) agent with FinRL. The state includes prices, holdings, and cash; actions change holdings through buying, selling, or…
The post asks why Alpha158 labels use different forward-return horizons in two implementations. It compares a label spanning the close at T+1 to the close at T+3 with a Qlib label spanning T+1 to T+2, then relates those choices to China’s T+1 stock-trading…
The article presents quantitative investing as a way to replace discretionary buy and sell decisions with signals from systematic models. It attributes common retail timing mistakes to fear, greed, and reactions to market sentiment, and says rules-based…
This research summary presents TS-Boost, an equity factor-selection framework designed for financial data with shifting cross-sectional patterns and low signal-to-noise ratios. Instead of pooling observations from different dates into one training set, it…
This discussion describes a model-stacking problem in a Chinese quantitative research platform. A user has trained and persisted twenty StockRanker models using the same factor set but differing in date ranges, data filters, or prediction horizons. The goal…
This French-language overview compares five open-source tools for automating cryptocurrency trading: OctoBot, FreqTrade, Hummingbot, Jesse, and Superalgos. It describes their broad use cases and relative strengths, including configurable strategies, cloud or…
The document describes a daily classification approach for detecting positive momentum, negative momentum, or normal conditions in Bitcoin, Ethereum, and Litecoin. It uses historical prices and technical indicators as features, then compares classifiers…
This overview surveys practical considerations for AI and quantitative trading, from data quality and model choice to backtesting, risk controls, and portfolio allocation. It highlights that historical results can mislead when models are overfit or…
This Korean-language project overview describes LumiBot, a Python framework for building trading strategies that can use ordinary rules, AI agents, or a combination. It presents a workflow that begins with a sample strategy and historical-data backtest, then…
This course-lecture summary introduces the role of planning and learned models in reinforcement learning. It identifies Dyna and Monte Carlo tree search as examples of algorithms that use models, and notes that the lecture is presented by research engineer…
This page summarizes two research topics from an overseas literature review. The first concerns systematic value investing: it describes examining common claims about the value effect, considering how diversified value strategies might be implemented more…
This configuration defines a Qlib workflow for predicting short-horizon CSI 300 stock returns with a temporal convolutional network (TCN). It uses Alpha158 features, filters a specified set of feature columns, applies robust cross-sectional normalization,…
This post summarizes a 2017 J.P. Morgan report on using large and alternative datasets in investment research. It groups alternative data into information generated by individuals, business processes, and sensors, with examples such as social media,…
This short discussion outlines data-related challenges faced by quantitative investment teams as they develop and test strategies. It describes quantitative investing as turning historical market behavior into data, using statistics and programming to…
This article introduces Monte Carlo methods through random sampling examples, contrasting an approach that can return a promising answer without guaranteeing the optimum with randomized search that keeps trying until it finds a valid solution. It illustrates…
The indicator combines a SuperTrend-style trailing line with a volume-weighted moving average and a k-nearest-neighbors classifier. The trend bands are centered on the volume-weighted average and offset using average true range. For its classifier, the…
This brief BigQuant Q&A addresses whether to exclude stocks listed on China’s ChiNext and STAR Market boards when building an AI trading strategy. Its central recommendation is to apply the exclusion during both model training and prediction, so the model is…
This study presents StockRanker, a supervised learning-to-rank method for selecting Chinese A-shares. It uses gradient-boosted trees to rank the market by expected future return. The input features are derived from seven basic price and volume series using…
The post raises a reproducibility question: two strategy templates reportedly use the same model, factors, parameters, and data, yet produce substantially different backtest results. The author asks which result is correct and whether one template contains…
Qlib is introduced as a modular platform for researching quantitative investment strategies with AI and machine learning. Its components are loosely coupled, so parts of the platform can be used independently. The architecture is organized into…
This document indexes 50 articles published in 2017, grouped into topics such as computer vision, recommendation, games, speech, language processing, and prediction. The list includes a stock-price forecasting tutorial alongside examples from other fields,…
This research overview introduces six broad model families used in investment work: neural, graphical, clustering and encoding, linear, tree-based, and ensemble methods. It describes their general strengths, such as nonlinear function fitting, relationship…
The post introduces machine learning as an increasingly used approach in quantitative investing. It describes two applications: forecasting market movements and selecting portfolios. It also points to potential benefits, including handling nonlinear…