# factrix

> Polars-native factor signal validator. Tests one factor. Screens a thousand.
> Source: https://github.com/awwesomeman/factrix
> Docs: https://awwesomeman.github.io/factrix/

## Pages

- [Get Started](https://awwesomeman.github.io/factrix/getting-started/install/): installation, quickstart, three-axis concepts
- [Concepts](https://awwesomeman.github.io/factrix/getting-started/concepts/): FactorScope / FactorDensity / DataStructure / MetricSpec explained
- [Quickstart](https://awwesomeman.github.io/factrix/getting-started/quickstart/): five-minute end-to-end example
- [Guides](https://awwesomeman.github.io/factrix/guides/): Panel vs timeseries, large-scale evaluation, custom metrics, choosing a metric
- [API — inspect_data](https://awwesomeman.github.io/factrix/api/inspect-data/): pre-flight data inspection
- [API — evaluate](https://awwesomeman.github.io/factrix/api/evaluate/): single-factor and multi-factor evaluation entry point
- [API — Multi-horizon](https://awwesomeman.github.io/factrix/api/multi-horizon/): multi-horizon evaluation recipes using evaluate_horizons + bhy(expand_over=)
- [API — EvaluationResult](https://awwesomeman.github.io/factrix/api/evaluation-results/): unified result schema, to_dict() / to_frame()
- [API — multi_factor](https://awwesomeman.github.io/factrix/api/multi-factor/): bhy() / bhy_hierarchical() / partial_conjunction() — FDR-corrected screening
- [API — stats](https://awwesomeman.github.io/factrix/api/stats/): BHY FDR plus Holm and Romano-Wolf FWER adjusted p-values, dependent-series bootstrap helpers
- [Reference — warning codes](https://awwesomeman.github.io/factrix/reference/warning-codes/): WarningCode enum
- [Reference — metric applicability](https://awwesomeman.github.io/factrix/reference/metric-applicability/): which metrics apply to which axis combinations
- [Full LLM reference](https://awwesomeman.github.io/factrix/llms-full.txt): single-fetch file covering concepts + API + usage patterns
