ml4t-case-study-pipeline
Filesystem and artifact-contract pattern for reproducible case studies. Use when organizing a research project for reproducibility and collaboration.
38,472 Skills
Search uses this published catalog; rankings and counts are snapshots.
Filesystem and artifact-contract pattern for reproducible case studies. Use when organizing a research project for reproducibility and collaboration.
Validate causal claims using DAG adjustment sets, bad-control detection, and refutation tests. Use when distinguishing genuine factor effects from confounded associations.
Systematic feature computation across multiple assets with group-aware operations. Use when computing technical or fundamental features for a panel of securities.
Build roll-adjusted continuous futures series without artificial price jumps. Use when backtesting futures strategies that span contract rollovers.
Export financial data in efficient columnar formats with schema enforcement. Use when persisting datasets for reproducible research or cross-pipeline sharing.
Prevent train-test contamination, target leakage, and temporal leakage. Use when splitting data, fitting preprocessors, or engineering features for time-series ML.
Define point-in-time tradeable universes with liquidity filters. Use when constructing the investable asset set that avoids survivorship and liquidity bias.
Decompose portfolio into factor, sector, and concentration exposures. Use when checking for unintended bets or risk concentrations.
Five families of financial features - momentum, mean-reversion, volatility, carry, and value. Use when designing a feature set to ensure coverage across complementary market dynamics.
Select informative features using IC ranking, mutual information, or RFE - always within CV folds. Use when reducing feature dimensionality before training.
Organize computed features in versioned storage with schema enforcement and point-in-time retrieval. Use when persisting features for reproducible ML experiments or sharing across pipelines.
Validate features before training - IC significance, stability, redundancy, and contamination checks. Use when auditing feature quality before model fitting.
Choose prediction horizon by analyzing IC decay, turnover cost, and feature-horizon alignment. Use when determining the optimal lookahead window for labels.
Measure predictive signal quality with IC, Rank IC, and IC_IR. Use when evaluating whether a feature has predictive power for returns.
Automated risk limits that halt trading when thresholds are breached. Use when deploying live strategies that need drawdown or loss-limit protection.
Extract latent factors from return panels using PCA, IPCA, or autoencoders with proper noise diagnostics. Use when reducing dimensionality or discovering risk structure.
Catalog captured:2026-11-21