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.
715 documents
This notebook turns registered NASDAQ-100 microstructure backtests into a strategy assessment without training models or rerunning backtests. It reads metrics and lineage from the registry, builds derived cohort and paired-comparison tables when needed, and…
This notebook applies a previously selected CME futures configuration to holdout predictions, using the established allocator, rebalance schedule, concentration rules, and transaction cost assumptions. Its central methodological point is to freeze those…
This notebook describes how to select a CME futures case-study configuration from registered validation backtests. It pools candidates across signal, allocation, and risk-management stages, then chooses the configuration with the strongest validation Sharpe.…
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 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…
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 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 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 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 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 fits ridge, lasso, and elastic net models to published firm characteristics for individual US stocks. The features span value, profitability, investment, size, momentum, reversal, risk, and liquidity. Walk-forward folds keep configurations…
This case study fits gradient-boosted trees to ETF return labels and compares them with a linear baseline that performed well under strong ridge regularization. It explains why trees can discover interactions, such as momentum behaving differently across…
This exploratory analysis profiles a historical US equities panel, explains its raw and adjusted price fields, and examines symbol coverage over time. It compares yearly counts of active symbols with first and last observations, showing why both entry and…
This notebook builds conditional-volatility features for an S&P 500 equity and options study using a GJR-GARCH model. Unlike fixed rules based on past prices, fitted features depend on their estimation window. The method therefore declares a refit schedule,…
This document demonstrates feature-store concepts directly with case-study Parquet data. A feature view specifies the entity key, event timestamp, usable lifetime, and feature columns. For model training, historical values are joined to the decision time at…
This analysis notebook explains how to assess a US equity strategy built from firm characteristics using registry results rather than new backtests. It reads uncertainty-aware performance statistics, ranks configurations on shared support, and uses paired…
This document describes a benchmark for database engines used with financial time-series data. It compares bulk writes, full scans, time-range queries, OHLCV aggregation, and as-of joins for trade and quote data. Its methodology treats client-side conversion…