• PyFixest
  • Getting Started
  • Function Reference
  • Tutorials
    • All Tutorials

    • OLS with Fixed Effects
    • Formula Syntax
    • Regression Tables
    • Difference-in-Differences
    • Quantile Regression
  • Guides
    • All Guides

    • Marginal Effects & Hypothesis Testing
    • Regression Decomposition
    • PyFixest on the GPU via JAX
    • PyFixest on the GPU via CuPy
    • Compare Stata & PyFixest
    • Variance Reduction in AB Tests
  • Explanation
    • When Are Fixed Effects Problems Difficult?
    • On Small Sample Corrections
    • Does PyFixest Match fixest?
  • Textbook Replications
    • All Textbook Replications

    • Causal Inference for the Brave and True
    • Causal Inference: The Mixtape
    • The Effect
    • Other Resources
  • Changelog
  • Contributing
  • Acknowledgements
  • Sprint 2026
  • llms.txt
  • Skills.md
Categories
All (5)
AB Testing (1)
Panel (1)

Guides

Practical recipes and deeper dives for PyFixest practitioners.

Task-oriented guides that show you how to accomplish specific goals with PyFixest.

Marginal Effects and Hypothesis Tests via marginaleffects
We can compute marginal effects and linear and non-linear hypothesis tests via the excellent marginaleffects package.
How much do Covariates Matter?
In regression analyses, we often wonder about “how much covariates matter?” for explaining the relationship between a target variable \(D\) and an outcome variable \(Y\).
PyFixest on professional-tier GPUs
PyFixest allows to run the fixed effects demeaning on the GPU via the demeaner_backend argument. To do so, you will have to install jax and jaxblib, for example by typing pip…
Translating Stata to PyFixest
This guide will focus on how to replicate the regression results you would get in Stata with the Python package pyfixest and assumes you know how to do things like install…
Variance Reduction in AB Tests
Panel
AB Testing

How to reduce the variance of estimators in AB tests and beyond using exogenous control variables via CUPED, linear regression, and fixed effects estimators.

No matching items