factrix 0.11.0

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?

Where factrix fits

factrix is the first Python framework to dispatch primary statistical tests by factor type — cross-sectional, event, and common factor each get the test that fits their data-generating process.

factor construction  →  factrix (verdict)  →  strategy construction  →  backtest  →  live trading
                            ▲ you are here

For each candidate factor factrix answers — is the predictive power real? — and corrects for multiple testing when you screen at scale. Kill fakes before they cost you a backtest.

Why factrix?

• Type-routed evaluation — Information Coefficient + Fama-MacBeth for cross-sectional factors; Cumulative Average Abnormal Return for events. Each type ships its own multi-metric diagnostic battery.
• Batch factor screening — screen hundreds of candidate factors with cross-test multiple-testing correction in a single API call.
• Financial statistics built in — autocorrelation-robust standard errors (Newey-West), overlapping-forward-return correction, persistent-predictor flagging (Stambaugh bias), false-discovery-rate control across batches (Benjamini-Hochberg-Yekutieli). Hand-rolled Information Coefficient loops typically run scipy.stats.ttest_1samp on the IC series and ignore the autocorrelation induced by overlapping forward returns.
• Polars-native — modern Polars alternative to the pandas-based alphalens.

factrix stops at the verdict — primary test plus diagnostic battery. It does not size positions, model slippage, optimise weights, or compose alphas; those belong to the later stages of the pipeline above.

Is factrix the right tool?

You want to…	Use this
Verdict on a factor (cross-sectional / event / common factor)	factrix
Screen many factors with multiple-testing correction	factrix
Backtest with positions / slippage / margin	zipline-reloaded, backtrader, bt, vectorbt, nautilus_trader
Optimise portfolio weights	skfolio, riskfolio-lib
Returns-level tear-sheet (P&L diagnostics)	pyfolio-reloaded, QuantStats
Familiar cross-sectional tear-sheet	alphalens-reloaded
End-to-end machine-learning pipeline	qlib
Deflated / probabilistic Sharpe today (commercial)	mlfinlab

Where factrix fits — full comparison →

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 fx
from factrix.preprocess import compute_forward_return

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

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

Multi-factor BHY screening

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

Single-asset (timeseries) fallback

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