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 reading note summarizes two quantitative research papers. The first applies the thermal optimal path method, a nonparametric approach motivated by statistical physics, to study how stock indices and their futures lead or lag one another over time. The…
This brief Chinese-language listing points to a lecture on model-free prediction in reinforcement learning. It identifies the topic as the relationship between prediction methods and Monte Carlo and temporal-difference algorithms, and names Hado van Hasselt…
The introduction presents a machine-learning framework for selecting securities for pairs trading. It frames pair discovery as a search-space problem: limiting candidates to securities in the same sector may exclude useful relationships, while searching…
The discussion examines a sequence-window feature pipeline for stock data. If rows are filtered before a rolling window is formed, dates can disappear; a nominal three-row window may then combine observations that are not three consecutive days. The answer…
The document explains fixed-width fractional differencing (FFD) as a way to make price series more suitable for statistical and machine-learning models while retaining some long-term dependence. It contrasts this with ordinary integer differencing, which can…
This short homework response defines quantitative investing as using scientific methods to seek profitable predictions. It identifies systematic analysis and statistical reasoning as strengths, then cautions that simulated results can create a false sense of…
This Python example outlines a deep reinforcement learning workflow for a stock portfolio using FinRL and Alpaca. It trains a PPO agent with ElegantRL on one-minute data for Dow Jones stocks, evaluates it on a short held-out date window, then retrains using…
This short note highlights that feature importance can differ across predictive models, naming StockRanker and XGBoost as examples. It also points to changing the number of leaf nodes and the learning rate as model-tuning choices. These observations can help…
MarketPredictor is described as an open-source MetaTrader 5 Expert Advisor that applies mathematical tools to historical prices and future scenarios. Its stated methods include Fast Fourier Transform analysis to identify dominant frequency patterns and Monte…
This article explains a GBDT-plus-logistic-regression method for classifying stock direction. Gradient-boosted decision trees capture nonlinear feature relationships; each tree maps observations to leaf nodes, which are encoded as indicator variables.…
A user reports that a template stock strategy fails during model training after replacing its full A-share universe with a manually chosen set of only a few stocks. The response recommends expanding the sample pool, explaining that the template is designed…
This report presents machine learning as a data-driven approach and distinguishes supervised learning, which maps labeled inputs to outputs, from unsupervised learning, which seeks structure in unlabeled data, and reinforcement learning, which learns actions…
This report introduces machine learning as data-driven learning and outlines its three broad forms: supervised, unsupervised, and reinforcement learning. It describes how computing advances, larger datasets, and accessible software libraries have helped make…
This brief overview introduces genetic programming as an evolutionary approach to generating quantitative investment factors. It describes starting with basic data inputs and mathematical operators, then iteratively evolving candidate expressions through…
This overview presents LumiBot as a Python framework for writing conventional rule-based strategies, AI-agent strategies, or combinations of the two. Its workflow supports running historical backtests before connecting to a broker, then running a strategy in…
The article discusses translating plain-language trading rules into Python strategies for the QMT platform. Its example is a moving-average crossover on a named Chinese equity, with an all-in position rule and adjusted price data. The author reports that an…
The document describes using genetic programming to discover stock-selection factors from intraday price and volume data. Factor expressions are represented as trees; successive candidate populations are produced through selection, crossover, and mutation.…
The document outlines an end-to-end portfolio construction framework that embeds a differentiable convex optimization layer within a neural network. Raw asset data passes through the network for feature extraction and combination, then an optimization layer…
This guide outlines six compact trading bot demos, each built around a single AI agent using plain-language instructions and built-in data tools. The examples include discretionary stock selection, market news, news sentiment, trend following, a…
This Chinese-language forum question explains two data objects that appear in AI strategy examples on the BigQuant platform. The ranker prediction object contains predictions passed to the backtesting module, and the strategy can read the data through the…
This tutorial introduces binary classification using logistic regression, with stock direction as its example. It explains encoding outcomes as two labels, interpreting the model output as a probability, and applying a threshold to produce a class. Logistic…
This example describes a daily stock selection process in which a research agent ranks large US stocks using recent prices, trends, and news. Bull and bear agents assess the same research from opposing perspectives, then a judge and trading agent select the…
This project plan outlines VeighNa’s intended 4.0 development, including support for Python 3.13, framework refactoring, improved logging and API bindings, updated packaging and code-quality tools, and expanded trading gateways and applications. It describes…
This brief listing identifies an educational quantitative trading strategy that uses linear regression to predict returns. It names the approach and its purpose, and indicates that an implementation is available in the associated platform environment. The…