OpenQuant Documentation

Documentation Principles

• Workflow-first navigation: documentation is organized by practical execution path, not internal project structure.
• Methodology traceability: all major workflows map back to AFML chapter concepts and caveats.
• Operational realism: setup, performance, and governance pages include commands and decision criteria used in production.
• Evidence over claims: examples, formulas, and failure-mode notes are explicit on each module page.

Recommended Reading Path

1. Quickstart
2. Prerequisites
3. Local Build Setup
4. Rust Core Workflow
5. Python Core Workflow
6. By AFML Chapter
7. Module Reference Index
8. Methodology and Leakage Controls

Core Workflow Lanes

Event-Driven Data and Labeling (AFML Chapters 2-3)

• Event bars and filters: reduce microstructure noise and avoid naive time-bar bias.
• Triple barrier and meta-labeling: explicit event horizon and side/size separation.

Validation and Diagnostics (AFML Chapters 4, 7, 8, 9)

• Purged and embargoed validation: leakage-resistant split policies.
• Feature diagnostics and tuning: substitution effects, orthogonalization, and disciplined search.

Backtesting and Risk (AFML Chapters 11-16)

• CPCV and synthetic backtesting for robustness.
• Portfolio and strategy risk metrics with explicit assumptions and caveats.

What Is Included In This V1 Institutional Baseline

• Setup and quickstart runbooks.
• Full module reference pages with APIs, formulas, examples, and notes.
• Indexes by workflow, AFML chapter, and module.
• Governance pages for leakage controls, reproducibility, and benchmark policy.

What Is Not Included Yet

• Formal audit evidence bundles.
• Change-control approval logs integrated with external governance systems.