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 forum post discusses a binning failure in a quantitative label-generation workflow. It attributes the reported non-unique bin edges to a weighted-binning step that attempts to divide a label into 20 groups. The preceding label expression clips values at…
This Chinese-language forum post reports a NameError in a deep neural network stock strategy tutorial: the variable holding stacked factor data is undefined when downstream code tries to read its date range. The shown workflow expects that data frame to…
This article adapts a machine-learning value strategy to Chinese A-shares. It trains an XGBoost regression model on accounting and company characteristics to forecast future price-to-book ratios. The gap between forecast and current price-to-book, scaled by…
This Chinese-language post reviews several strategy ideas in the context of a reported shift from leading stocks toward reversal patterns. It describes a multi-factor approach that looks for rebound or reversal behavior after attention-grabbing stocks pull…
The report introduces generative adversarial networks (GANs) for quantitative finance. A generator creates synthetic market data while a discriminator tries to distinguish it from real data; alternating training aims to make the generated series resemble the…
A BigQuant community post reports that an XGBoost template completes historical backtesting but fails when submitted for simulated trading. The traceback shows the failure occurring while training data is prepared: the code calculates ranking group sizes…
The document presents TS-Boost as a machine-learning framework for equity factor selection, motivated by two challenges in financial data: changing distributions across time and low signal-to-noise ratios. Instead of pooling observations from many dates into…
This research note explains how one-dimensional convolutional neural networks can extract local patterns from financial time series, and how ResNet skip connections help address performance degradation as networks deepen. Its proposed stock-ranking setup…
The strategy applies K-Means clustering to daily index features, such as recent average volatility, MACD, and average trading volume, to group trading days into market regimes. With two clusters, it compares each day’s feature vector with cluster centers,…
This document explains how to use the FuzzyNet library in MQL5 to build fuzzy inference systems of the Mamdani or Sugeno type. It outlines a workflow: define input and output variables, assign linguistic terms and membership functions, create rules, provide…
This overview describes alpha as a return opportunity identified by studying historical data and relationships among market, company, macroeconomic, and text information. It outlines a research cycle: develop an economic rationale and measurable signal,…
This announcement outlines a workshop on applying Kronos, a language-model-style system designed for financial price series, to CTA research. Its central idea is to represent candlesticks as tokens so a Transformer can learn patterns across sequences. The…
This configuration describes a Qlib workflow for training a TCTS model on the CSI 300 universe using Alpha360 features. The data span 2008 to 2020, with training through 2014, validation over 2015–2016, and testing from 2017 to mid-2020. Feature processing…
The article answers practical questions about machine-learning stock strategies. It says the platform separates training and test periods by time, with the test interval serving as the backtest, and describes a setup without a separate validation set. It…
This report describes a Chinese equity strategy that predicts the future effectiveness of seven style factors and adjusts their portfolio weights over time. It uses historical information coefficients, macroeconomic data, and market variables as model…
The document describes a candlestick fitness rule intended for use in coding high-frequency trading algorithms based on population optimization. It labels a candle unfit when its body does not meet a default proportion of the relevant upper or lower wick…
This brief Chinese-language support note explains a prediction-data problem traced to missing values in a rolling price-range factor. The example factor compares a short moving average of the high-low range with a longer moving average. According to the…
This brief introductory response defines quantitative investing as finding patterns in historical data, translating them into a model and strategy, and using software to execute trades according to the resulting rules. It highlights discipline as a benefit:…
This excerpt summarizes a Chinese weekly research note on machine learning and commodity trading advisor strategies. It reports recent results for a neural network strategy on the CSI 500, a commodity futures strategy, and a commodity strategy combining…
This document explains how to remove ChiNext stocks from both the training and test portions of a visual AI stock-selection strategy on BigQuant. It instructs users to insert a shared “filter market” module before the missing-data handling module in each…
The study describes a machine-learning approach for classifying the optimal rebalance frequency of a two-asset portfolio strategy. It uses minute-level prices for 50 Binance crypto assets from 2022 and 2023, forms asset pairs, and groups them by correlation.…
The document introduces Qlib’s online-serving components for applying trained models to current market data. It describes a workflow that can produce predictions in live conditions and support real trading based on those predictions. The named components are…
This presentation transcript surveys quantitative trading practices described for overseas markets. It explains how nontraditional inputs such as news sentiment, ESG reports, satellite imagery, and weather can be converted into structured signals, while…
This article describes using recurrent neural networks to predict the direction of the next price change from high-frequency order-book data. The study uses data from Nasdaq stocks and keeps observations where prices change, reducing the input sequence. It…