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
This guide outlines an end-to-end quantitative research workflow using Qlib. It describes installing the library, preparing Chinese market data from public sources, and running an example LightGBM configuration with the qrun tool. That workflow combines…
This brief BigQuant platform discussion asks how to implement complex methods for combining investment factors and whether suitable operators or code are available. The reply first asks what kind of combination the user has in mind, then says that modules…
This note compares DeepAlpha deep neural network and convolutional neural network stock selection experiments in a small-cap universe. The author describes a daily strategy that buys five stocks and sells them the next day, using 57 factors drawn from 19…
This competition FAQ explains how submitted equity factors are reviewed and how model training should be separated from evaluation. It warns that training on the test dataset, or fitting a model on a portion of data and then using that same portion to…
This post describes an author’s deep-learning model built with market data and shared through open APIs, alongside links to simulation and backtest examples. It discusses stock selection from large-cap and mid-cap Chinese equity universes and says users can…
This overview introduces prediction as either regression, which estimates numeric outcomes, or classification, which predicts categories. It groups six common algorithms into linear models, tree-based models, and neural networks. Linear regression and…
The document asks whether feature importance can be obtained from logistic regression, including Lasso-regularized models. Its reply contrasts tree-based importance measures with coefficient-based models and notes that a platform with scikit-learn installed…
This brief BigQuant forum post asks how to build one feature from another when the first feature is a moving-average comparison. The desired first feature assigns a positive value when the five-day average exceeds the ten-day average and a negative value…
This paper examines whether deep neural networks can learn delta hedging strategies for options when the idealized assumptions behind Black–Scholes—such as continuous trading and no transaction costs—do not hold. It compares recurrent neural networks,…
The document briefly describes CryptoTrend 1.00, an automated expert advisor for Bitcoin trading, and says it was optimized for 2024. Its stated adaptive component tracks counts of winning and losing trades and uses those statistics to adjust a trend-entry…
This configuration defines a Qlib workflow that trains a LightGBM model on Alpha158 features for the CSI 500 universe. It assigns data from 2008 through mid-2020 to training, validation, and test segments, with the fit period ending in 2014. The model uses…
This report describes adapting a platform’s deep neural network template for stock selection. The model uses 98 price and volume factors and three fully connected layers. The author uses a Calmar ratio label, about ten years of training data, and a universe…
This document reviews a dozen papers presented around AAAI 2022. Among them, the most relevant to quantitative research is a method for adapting to predictable concept drift in time series. DDG-DA learns weights for resampling historical observations so that…
This overview surveys twelve papers presented at AAAI 2022, spanning machine learning, vision, language, and online optimization. Its most direct trading-related contribution is DDG-DA, a method for predictable concept drift in time series. Rather than only…
This page introduces a lecture on approximate dynamic programming from a reinforcement learning course. It identifies the subject as the performance of approximate algorithms and names researcher Diana Borsa as the presenter. The topic is relevant to…
This platform discussion concerns how to use computing resources for hyperparameter search in quantitative models. The author compares low-resource and higher-resource single-machine environments, reporting little speed improvement under default settings and…
This Chinese Q&A collection covers practical topics in quantitative stock research and strategy deployment. It explains cross-sectional versus time-series ranking, preprocessing such as missing-value and outlier handling, neutralization and normalization,…
The document introduces Condor, an open-source interface for managing trading bots and autonomous agents through natural language, Telegram, or a web dashboard. Its central design principle separates probabilistic reasoning by a large language model from…
GeneralPtNN is presented as a redesign intended to support both time-series and tabular datasets through a common PyTorch workflow. The stated approach is to keep the workflow configurable and change the network and dataset classes when moving between data…
This article describes a supervised learning approach to identifying stocks at elevated risk of severe relative price declines before events such as bankruptcy or credit downgrades. It combines economically motivated company and market features, including…
This review surveys machine learning approaches to stock forecasting, including support vector machines, random forests, nearest-neighbor and Bayesian classifiers, neural networks, and recurrent models. It also describes research examples that combine…
This document outlines a stock selection approach using daily price amplitude above 1, a positive weekly MACD histogram, and an arc-shaped price pattern. It interprets the amplitude condition as a volatility filter, the weekly indicator as evidence of upward…
This introductory overview explains common concepts from neural networks, convolutional neural networks, and recurrent neural networks. It describes neurons, weights, biases, activation functions, layers, and the forward and backward passes used to train a…
The author reflects on the difficulty of building a strategy that remains profitable across bull and bear markets. They identify detecting changes in market style as a promising research problem and describe the challenge of interpreting AI models whose…