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 article describes a stock selection experiment in which a machine learning model used only days since listing. Its reported backtest showed positive returns, though with high volatility. The authors attribute performance partly to a rising market during…
This example describes an AI agent workflow for ranking and trading large-cap stocks. A researcher ranks the stock universe, bull and bear agents present opposing cases, and an interpreter turns the debate into target account weights. A trader agent reads…
This page summarizes a lecture on multi-step and off-policy algorithms in reinforcement learning. It identifies the subject as methods that learn from sequences of rewards across multiple steps while handling data generated by a policy different from the one…
This short assignment response defines quantitative investing as turning a personally developed set of screening and trading rules into a computer-executable process. It presents rule design and implementation as the core idea: formulate conditions for…
This submitted assignment defines quantitative investing as making investment decisions with mathematical and statistical analysis, historical data, and computing. It describes several practical advantages: processing large datasets, applying rules…
The report describes an AI-based score for assessing intraday trading opportunity in Chinese stocks and broad indices. Using 20 trading days of one-minute returns, it compares observed returns with randomly constructed return sequences. A higher score…
The document outlines a trade filter based on k-nearest neighbors. It stores vectors describing past trades or market conditions, labels outcomes as profitable or unprofitable, and uses Euclidean distance to find nearby examples for a new situation. The…
This Chinese Q&A explains why a genetic programming factor-mining workflow can report factors that pass a test yet return an empty factor collection. The answer says the run log shows that the global fitness test did not pass, so no factors were retained. A…
This guide lists reasons a submitted quantitative factor may fail to produce an evaluation score. It recommends checking factor-analysis output on the designated dataset before submission and verifying that a machine-learning model completed training and…
This live strategy dashboard describes a multi-asset perpetual futures system that combines factor signals, assigns target exposures across contracts, and submits passive limit orders. The listed seed factors cover reversal, momentum, premium, funding, and…
This introductory explanation describes how a convolutional neural network turns an image into a class prediction. A filter slides across local pixel regions to produce feature maps; filter count, stride, and zero padding affect the maps’ depth and…
BreakRevertPro is described as an automated trading system that combines breakout and mean-reversion approaches. Its trade identification uses statistical distributions, including Weibull, Poisson, and exponential models. The listed features also include…
This short platform guide explains how to limit the prediction or validation stock universe in a visual AI stock selection workflow to a chosen concept sector. It instructs users to insert a shared module for selecting stocks from a specified concept sector…
This discussion explains when stock universe filters should apply to both model training data and prediction data, and when they might be applied only at trading time. Its general recommendation is to use the same data processing in both sets. Filters that…
This brief Chinese-language assignment submission argues that factor quality matters more than the choice of predictive model: a strong factor is presented as the foundation of a strong strategy. It says the author ran a linear-regression strategy because…
The document discusses constructing supervised-learning targets from daily Chinese stock data. Its SQL example derives a forward return from later prices, clips extreme values using the first and ninety-ninth percentiles, and selects dates, instruments, and…
This report introduces six ways to interpret machine-learning models and applies them to an XGBoost stock-selection model. It discusses feature importance, individual conditional expectation (ICE), partial dependence plots (PDP), a surrogate decision tree…
The document introduces support vector machines, outlining hard-margin classification for linearly separable data, soft margins for data with outliers, and kernel methods for nonlinear boundaries. It then applies an SVM with a radial basis function kernel to…
This Qlib configuration defines a Chinese equity workflow using the CSI 300 universe and its associated benchmark. It prepares Alpha360 features with robust score normalization and missing-value filling, while labels are rank-normalized after missing labels…
This annual review summarizes VeighNa’s 2025 development, including its tenth year as an open-source project and the release progression from version 4.0.0 to 4.3.0. It describes vnpy.alpha, a workflow for machine-learning research and trading that covers…
These reading notes survey a Python data-analysis workflow, from loading files and cleaning observations to transforming, modeling, and presenting data. They introduce NumPy arrays and vectorized operations, pandas Series and DataFrames, and plotting with…
This Freqtrade hyperoptimization module searches for rule-based long entries and exits using a catalog of price, volume, volatility, trend, and momentum series. For each rule, the search space selects an indicator, a second indicator or comparison target, a…
The document describes a neural network that predicts Bitcoin prices ahead but produces a curve that resembles observed prices with a persistent lag. The discussion examines possible data alignment and model behavior, then reports that removing an unrealized…
This exchange concerns tools for investigating the Dragon-King hypothesis, which proposes that some financial crises and crashes may be predictable rather than behaving like ordinary extreme events. The question asks for software or code to test that idea.…