Pingouin is a free, open-source statistical package written in Python 3 and based mostly on Pandas and NumPy. It includes statistical tests, regression functions, Bayesian analysis, and scripting support, with CSV and XLSX data import through Pandas. Correlation methods include robust, partial, distance, repeated-measures, and pairwise approaches. Modeling functions cover linear and logistic regression, mediation analysis, and multivariate tests. The package also provides effect sizes, power analysis, and parametric or bootstrapped confidence intervals for effect sizes or correlation coefficients. Its plotting functions include Bland–Altman, Q-Q, and paired plots. Several functions, including ANOVA, pairwise tests, correlations, and mediation analysis, can be called as Pandas DataFrame methods. Installation is available through uv, pip, or conda. The documentation says the project is under development, warns that bugs may exist, and advises checking results with other statistical software; it provides no warranty. The GitHub repository identifies the license as GPL-3.0. Questions can go to GitHub Discussions and bug reports to the repository.
Who it is for
Pingouin suits Python users who need statistical tests, modeling, or analysis integrated with Pandas workflows. Users should account for the project’s development warning and verify results with other statistical software.
What is good
- Free and open-source under GPL-3.0.
- Offers linear and logistic regression functions.
- Provides power analysis and effect sizes.
- Some analyses work as Pandas DataFrame methods.
- Installable through uv, pip, or conda.
What to know first
- Project says it remains under development.
- Documentation warns bugs may exist.
- The project provides no warranty.
Verdict
Pingouin offers a broad set of statistical functions within a Python and Pandas-oriented package. Its own development warning makes independent result checking important.
Get started with Pingouin
- Install Python 3.11 or later.
- Install Pingouin with uv, pip, or conda.
- Import .csv or .xlsx data through Pandas.
- Use Pingouin functions or call supported methods directly on Pandas DataFrames.
- Consult GitHub Discussions for questions or the GitHub repository to report bugs.
What the free plan stops at
The package is free, with no paid plans or usage limits stated. The documentation says it is tested for Python 3.11 and later; the project warns that bugs may exist and recommends checking results with other statistical software.
Questions about Pingouin
Is Pingouin free?
Yes. The Free and open-source plan is free, and no paid plans or usage limits are stated.
Is Pingouin open source?
Yes. Its GitHub repository identifies the license as GPL-3.0.
What data formats can it import?
It can work with .csv and .xlsx data through Pandas.
Which Python version does the documentation say it is tested for?
The documentation says Pingouin is tested for Python 3.11 and later.
Who maintains Pingouin?
The documentation says it was created and is maintained by Raphael Vallat, mostly in his spare time.
Where can users ask questions or report bugs?
The documentation directs questions to GitHub Discussions and bug reports to the GitHub repository.
Pingouin plans and pricing
All plansCompared on statistical analysis software
- Statistical tests
- Yes
- Regression models
- Yes
- Scripting support
- Yes
- Bayesian analysis
- Yes
- Data import formats
- .csv, .xlsx via Pandas
