Subjects (AFML)
Conceptual map from subject to OpenQuant modules.
OpenQuant-rs Documentation
Subject-first documentation aligned to core AFML domains so navigation follows quant research workflows, not project notes.
Subject Dividers (Used Across All Core Pages)
| Subject Divider | AFML Orientation | What You Find |
|---|---|---|
| Event-Driven Data and Labeling | Ch. 2-3 | Bars, filters, triple-barrier labeling, sample events. |
| Sampling, Validation and ML Diagnostics | Ch. 4, 7 | Sequential bootstrapping, purged CV, feature importance, model fingerprinting. |
| Position Sizing and Trade Construction | Ch. 10 | Probability/dynamic/budget/reserve bet sizing, ETF and futures roll helpers. |
| Portfolio Construction and Risk | Ch. 16 + risk overlays | CLA, HRP, HCAA, ONC, mean-variance family, VaR/ES/CDaR, backtest stats. |
| Market Microstructure, Dependence and Regime Detection | Microstructure + diagnostics | Kyle/Amihud/Hasbrouck lambdas, entropy features, codependence, SADF/CUSUM-style regimes. |
Where To Read
Module Pages
One page per module with formulas, API references, and code examples.
API by Subject
Primary functions and public types grouped by the same subject dividers.
Examples by Subject
Working snippets organized with the same domain taxonomy.
Engineering and Release Tracks
Performance and Publishing stay separate from the subject docs so research navigation remains clean.
Validation Commands
# Fast CI equivalent
cargo test --workspace --lib --tests --all-features -- --skip test_sadf_test
# Long-running SADF hotspot (explicit slow path)
cargo test -p openquant --test structural_breaks test_sadf_test -- --ignored
# Benchmarks and threshold check
cargo bench -p openquant --bench perf_hotspots --bench synthetic_ticker_pipeline
python3 scripts/collect_bench_results.py --criterion-dir target/criterion --out benchmarks/latest_benchmarks.json --allow-list benchmarks/benchmark_manifest.json
python3 scripts/check_bench_thresholds.py --baseline benchmarks/baseline_benchmarks.json --latest benchmarks/latest_benchmarks.json --max-regression-pct 25