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.
1,124 documents
This notebook applies one monthly ETF momentum and yield-curve regime strategy in two ways: vectorized return arithmetic and a sequential account simulation. Both use the same target weights, assets, dates, and total trading cost, allowing the comparison to…
This document explains a five-gate process for evaluating nine trading case studies, from signal quality through validation, transaction costs, holdout performance, and evidence quality. The gates are cumulative, so a strategy reaches the final stage only if…
This notebook walks through a deployment rehearsal for a crypto funding-rate direction model. It trains a three-class LightGBM model on historical Binance-derived perpetual data, then fetches live hourly bars and funding rates from OKX, aggregates data to…
This notebook describes fitting a long short-term memory network to one-minute NASDAQ-100 microstructure features to predict forward returns at several horizons. The model processes a trailing sequence step by step, learning which earlier observations to…
This notebook compares volatility measurements, fits a heterogeneous autoregressive (HAR) model, and estimates roughness using Hurst exponents. It first contrasts intraday realized variance with daily estimators, separating overnight returns from session…
The notebook introduces a conditional autoencoder that extends instrumented principal component analysis by replacing its linear mapping from stock characteristics to factor exposures with a neural network. It retains the exposure-times-factor-return…
This utility module builds tables and figures for evaluating predictive models across validation and holdout data. Its diagnostics include confidence intervals around information coefficients, validation-to-holdout decay, selection-adjusted backtest…
This notebook presents a workflow for acquiring and maintaining financial time series with a unified manager, predefined or custom symbol universes, and partitioned Parquet storage. It covers single and parallel multi-symbol fetching, loading data into…
The chapter explains classical approaches to generating synthetic financial data, covering continuous-time price processes, GARCH volatility, bootstrap resampling and model comparison. It describes how models such as geometric Brownian motion,…
This document describes the architecture of a neural policy network that maps asset-level time series and context into bounded portfolio risk weights. Its per-asset backbone combines context embeddings, feature-wise modulation, variable selection, an LSTM,…
This exploratory analysis tests whether a twelve-month momentum signal, omitting the latest month, is associated with next-month returns across a 100-ETF universe. It builds a monthly panel from adjusted closing prices, ranks eligible ETFs into quintiles…
This notebook compares portfolio sizing methods for leading S&P 500 equity signal configurations. It first selects each configuration’s strongest full-coverage checkpoint using validation Sharpe, then applies alternative allocators—including score weighting,…
This document records a multi-stage forecast of whether the Federal Reserve will raise rates during 2026. Three agents search for evidence and produce probabilities, followed by debate and a supervisor update. Their views differ: two assign a low probability…
This notebook runs the NLinear member of a declared three-model sequence-learning population for S&P 500 options. It resolves the requests and checkpoints for the full population before fitting this member, so later notebooks can execute the LSTM and…
This notebook describes the holdout prediction stage of a US firm-characteristics study. It takes the model configuration selected using validation results, reconstructs its training specification and checkpoint, then fits it again using data that ends…
This configuration note explains why an L1-penalized, three-class logistic regression uses SAGA instead of scikit-learn’s liblinear solver. It cites multiclass compatibility in newer scikit-learn versions and reports that SAGA completed much faster in…
The notebook describes an expanding-window double machine learning analysis of whether 12-to-2-month momentum predicts next-month US firm returns after adjustment for beta, idiosyncratic volatility, market capitalization, and variance. It fits nuisance…
This notebook presents a pipeline for extracting structured corporate events from SEC 8-K filings and loading them into a Neo4j knowledge graph. A language model processes filings in batches to produce event records with subjects, relations, objects, and…
This notebook compares Optuna hyperparameter searches for XGBoost, LightGBM, and CatBoost on a firm-characteristics dataset. Each library receives its own search study, using a tree-structured Parzen estimator, median pruning, early stopping, and persistent…
This notebook compares behavior cloning and inverse reinforcement learning (IRL) using demonstrations from a time-weighted average price (TWAP) execution rule. Behavior cloning maps observed market states directly to actions, while IRL fits a linear reward…
This notebook applies double machine learning to examine whether the implied-minus-realized volatility spread is associated with subsequent equity returns after adjustment for observed confounders. It identifies realized volatility, equity momentum, and…
This demonstration checks whether one dual moving-average crossover implementation produces matching signals in backtest and live engines. Both engines receive the same prepared ETF bars and use the same strategy parameters and fill convention, so the…
This training utility describes an optimization loop for a DeePM-style model. It uses AdamW to update model parameters from a robust Sharpe-oriented loss, passing forward returns, volatility scaling, masks, transaction costs, and configured burn-in and…
This notebook examines tick-level AlgoSeek trade and quote data for AAPL during the March 16, 2020 market crash. It filters to regular trading hours, separates trade prints from NBBO updates, and studies intraday activity, spreads, trading sizes, venue…