factrix 0.8.0

Polars-native factor signal validation toolkit for quantitative finance

Tests one factor. Screens a thousand.

A polars-native factor validator. It answers the core question — Does this factor possess predictive edge?

Installation

pip install factrix
# or
uv add factrix

See the installation guide for version pinning and development setup.

Typical usage

Single factor — IC evaluation

import factrix as fl
from factrix.preprocess.returns import compute_forward_return

raw   = fl.datasets.make_cs_panel(n_assets=100, n_dates=500, ic_target=0.08, seed=2024)
panel = compute_forward_return(raw, forward_periods=5)

cfg     = fl.AnalysisConfig.individual_continuous(metric=fl.Metric.IC, forward_periods=5)
profile = fl.evaluate(panel, cfg)

print(profile.verdict(), '| primary_p =', round(profile.primary_p, 4))
print(profile.diagnose())   # WarningCode / InfoCode list

Multi-factor BHY screening

profiles  = [fl.evaluate(p, cfg) for p in [panel_a, panel_b, panel_c, panel_d, panel_e]]
survivors = fl.multi_factor.bhy(profiles, threshold=0.05)

Single-asset (timeseries) fallback

cfg     = fl.AnalysisConfig.individual_continuous(metric=fl.Metric.IC, forward_periods=5)
profile = fl.evaluate(single_asset_panel, cfg)  # mode auto-switches to TIMESERIES
print(profile.stats.get(fl.StatCode.TS_BETA))

Documentation

• Get Started — install, quickstart, three-axis concepts
• Guides — PANEL vs TIMESERIES, BHY batch screening, choosing a metric
• Reference — applicability tables, formulas, statistical methods
• Development — architecture, contributing

License

Released under the Apache License 2.0.