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 evaluates NLinear, a simple sequence model that subtracts the latest observed level from a lookback window and maps the resulting sequence to a forecast with a linear transformation. It fits the ETF model population on walk-forward folds and…
This case study fits linear models to predict returns and direction in crypto perpetual futures. Its feature matrix is dominated by variations of the perpetual premium, the gap between the contract and spot price that drives funding payments. It contrasts…
This notebook presents a conditional autoencoder for equity returns in which a neural network maps stock characteristics to nonlinear factor loadings, while another network extracts contemporaneous latent factors from characteristic-managed portfolio…
This notebook measures how transaction costs affect a fixed CME futures strategy configuration. It selects the reported configuration through a shared validation-stage process, carries its risk overlay into the cost runs, and applies a declared grid of…
This notebook explains how to rebuild a limit order book from DataBento market-by-order messages for a single NASDAQ symbol and trading day. It models each order, aggregates orders into price levels, and maintains separate bid and ask sides. Correct message…
This notebook compares ways to hedge a short European call when rebalancing is discrete and trades incur proportional costs. It defines self-financing terminal P&L, then trains a neural hedger to minimize expected shortfall under Heston simulated prices. The…
This notebook compares six long-only ETF allocation methods designed to reduce reliance on unstable estimates. It applies Ledoit-Wolf covariance shrinkage to estimators that use covariance, and contrasts mean-variance maximum Sharpe with minimum variance,…
This notebook compares single-objective hyperparameter tuning with a multiobjective search for LightGBM prediction models. The baseline maximizes validation information coefficient (IC). The NSGA-II search instead maximizes IC while minimizing normalized…
This chapter synthesizes nine case studies that take machine-learning signals through portfolio construction, trading costs, risk overlays, and frozen holdout evaluation. It treats each case study’s progression as the unit of analysis instead of ranking…
This assessment traces a single US equities strategy from a frozen validation backtest set to its holdout evaluation. It applies a deterministic rule: choose the candidate with the highest validation Sharpe, breaking ties by backtest hash. Registry records…
This notebook runs a selected NASDAQ-100 microstructure configuration on its registered holdout predictions. The model, allocator, concentration, rebalance schedule, risk overlay, and cost assumptions are inherited from earlier work and applied unchanged;…
This notebook compares ways to size positions in a US equities panel while holding the model, checkpoint, rebalance dates, and selected stocks fixed. Prediction-based methods scale capital by forecast magnitude or interval uncertainty; inverse volatility and…
This notebook explains how to select one strategy from a fixed set of US equity backtests and assess the resulting strategy on a separate holdout period. It validates that candidates share the required data and protocol identities, then ranks them by…
This notebook demonstrates an operational workflow for connecting a shared backtest and live strategy to an Interactive Brokers paper-trading session. It checks account identity and state, requests historical bars to initialize indicators, subscribes to…
This notebook compares three linear time-series models, a Transformer encoder, and parameter-free forecasts on daily SPY returns. The linear approaches map a historical window to a multi-day forecast, with variants that separate a smooth component from its…
This notebook presents a multi-round forecasting debate between bull and bear roles. Each side argues for a higher or lower probability of an event, sees the other side’s prior argument in later rounds, and reports a probability and supporting evidence. The…
This notebook examines whether gradient-boosted trees can find nonlinear relationships in NASDAQ-100 microstructure features that a linear model may miss. It focuses on the possibility that order-flow imbalance predicts returns differently depending on the…
This notebook explains an operator-style research agent for iterating on quantitative case studies. Instead of limiting the model to predefined forecasting actions, it gives it general tools for files, shell commands, registry queries, and data inspection.…
This feasibility analysis checks whether a daily long-short equity ranking strategy can be researched with the available US stock panel. It examines point-in-time universe construction using price and trailing turnover thresholds, compares proportional…
This notebook refits the configuration selected by earlier validation stages using pre-2021 history, then publishes predictions for a 2021 holdout. It retrieves the chosen configuration from a recorded candidate set or applies the same ranking rule when that…
This analysis introduces option-chain structure and examines a 2020 slice of S&P 500 options for eight underlyings. It explains moneyness, intrinsic and time value, Greeks, implied volatility, and the information represented by volatility skew and term…
This notebook implements an ESG headline workflow that selects news with keywords, assigns each selected headline to an environmental, social, or governance category, and scores sampled headlines with FinBERT sentiment. It measures the selected pool’s…
This notebook audits an annual-report corpus before it is indexed for financial research. It measures the delay between a fiscal period end and the public filing date, showing why point-in-time applications must use the publication date. It also identifies…
This notebook screens financial and model-based features for their relationship with the next eight-hour return of crypto perpetual contracts. For each settlement, it computes cross-sectional Spearman correlations across eligible contracts, then summarizes…