Knowledge library

This evaluation notebook screens financial and model-based features for their relationship to the next eight-hour return of crypto perpetual futures. It builds the evaluation sample from walk-forward validation windows, keeps the final holdout outside the…

This NASDAQ-100 case study investigates whether 15-minute microstructure signals can survive trading costs. The workflow evaluates order flow, quote staleness, and relative spreads, then compares a full-universe strategy with one restricted to…

This notebook checks whether a daily US equity ranking strategy has enough usable data for a fair evaluation before fitting any model. It explains how to form a point-in-time universe with historical constituents, using printed prices and prior-session…

This notebook studies how regularization affects a CME futures model built from correlated feature families, including carry, rolling Sharpe, momentum, and volatility measures. It compares Ridge, Lasso, and ElasticNet: Ridge shrinks coefficients while…

This notebook applies configured stop-loss, trailing-stop, and time-exit rules to the strongest validation-ranked signal or allocation strategy for each futures return horizon. Each rule is assessed on its own parent strategy, with parameters fixed in…

This notebook runs a previously selected US equities configuration on holdout predictions. The selected model, predictions, allocator, concentration, risk overlay, rebalance cadence, and transaction cost assumptions are inherited from earlier stages. Its…

This notebook describes adapting Diffusion-TS to generate synthetic ETF return sequences. Its denoiser predicts the clean sequence as a sum of polynomial trend, Fourier seasonal, and residual components, while a combined time-domain and Fourier-domain loss…

The chapter argues that repeatable research discipline is a more durable source of trading advantage than any individual strategy. It presents a workflow built around falsifiable hypotheses, out-of-sample evaluation, and corrections for multiple testing to…

This document presents a cost sensitivity analysis for a fixed ETF momentum strategy. It holds the universe, weights, schedule, and simulator constant while rerunning the backtest over a range of per-leg fees. The resulting curve shows how Sharpe ratio and…

The document explains why TabM is used to model perpetual funding panel data: its shared neural backbone combines all input features, while nonlinear layers can represent interactions. Unlike a tree that chooses one feature at each split, this design can use…

This notebook models institutional equity holdings from SEC 13F filings as a graph linking managers to securities, with sector relationships for aggregation. Its schema retains both the quarter-end report date and the later filing availability date: the…

This notebook applies a previously selected ETF configuration to its reserved holdout window. It derives the matching holdout predictions from the selected training identity and checkpoint, then uses the strategy's existing concentration, allocation, risk…

This evaluation workflow screens financial and model-based features against forward returns for a NASDAQ-100 intraday panel. It first reserves the final test period, excluding observations whose forward labels would settle within that window. For each…

This notebook demonstrates how a broker wrapper can reject unsafe orders using order-size and position limits, order-rate caps, asset restrictions, and a persistent kill switch. A synthetic broker makes the decisions observable: attempted orders are either…

This analysis estimates whether skip-recent momentum is associated with a causal change in ETFs’ forward returns, rather than asking how well momentum predicts returns. Double machine learning models the outcome and the treatment using declared confounders,…

This benchmark-generation method creates an equal-weighted daily return series from the declared assets in a case study. It first takes each symbol’s last close for each session, then calculates close-to-close returns and averages the available returns…

This document describes a daily US equities dataset sourced from NASDAQ Data Link’s Wiki Prices archive. It covers thousands of companies from 1962 through March 2018 and includes delisted firms, making it useful for historical research that needs to reduce…

This analysis profiles a fixed universe of 100 ETFs across nine asset groups. It checks coverage by fund and year, confirms that the price panel matches the group configuration, reviews nulls and OHLC ordering, and compares trading volume across groups. It…

This notebook studies gradient-boosted models for forecasting returns and direction in crypto perpetuals. Its grid varies tree capacity and loss function, including squared-error, absolute-error, and Huber objectives, while scoring models at multiple points…

This document presents the PatchTST neural-network backbone for forecasting multivariate time series. It divides each input series into overlapping patches, projects each patch into a learned feature space, adds positional encodings, and processes the…

This document explains how to measure returns around events such as signal triggers, earnings announcements, or macro releases. Its worked example uses momentum breakouts in liquid ETFs. For each event, a market model fitted to a pre-event estimation window…

This notebook estimates whether the premium z-score of a crypto perpetual affects subsequent eight-hour returns when three declared drivers—price volatility, funding rate, and average premium deviation—are held fixed. Because treatment is continuous, the…

This notebook presents an unconditional Sig-Wasserstein GAN for generating financial time series. It replaces a learned real-versus-generated discriminator with a distance based on expected path signatures, mathematical summaries of a path’s sequential…

This notebook uses daily ETF returns to explain how GARCH models capture volatility clustering and turn it into a session-by-session feature. It first uses visual diagnostics and an ARCH-LM test to assess whether squared returns depend on their own lags. A…