maximtrp/scikit-posthocs

Multiple Pairwise Comparisons (Post Hoc) Tests in Python

What it solves

It fills gaps in the Python statistical ecosystem by providing a comprehensive suite of post hoc tests for pairwise multiple comparisons. These tests are used in statistical data analysis to determine exactly which group levels differ from one another after a statistically significant result has been obtained from an ANOVA test.

How it works

The package integrates with NumPy arrays and Pandas DataFrames to perform fast computations. It offers a wide array of parametric and non-parametric tests, including omnibus tests, tests for factorial and block designs, and comparisons against a control group. It also includes tools for outlier detection and p-value adjustment to minimize type I and II errors.

Who it’s for

It is designed for statisticians, data analysts, and researchers who use Python for their data analysis workflows.

Highlights

  • Extensive Test Library: Includes parametric tests (e.g., TukeyHSD, Games-Howell), non-parametric tests (e.g., Dunn, Conover), and omnibus tests.
  • Design Support: Specialized tests for both factorial and block design data.
  • Data Integration: Native support for Python lists, NumPy arrays, and Pandas DataFrames.
  • Visualization: Built-in significance plots (heatmaps) to visualize p-values.
  • Analysis Tools: Features outlier detection (e.g., Grubbs, ESD test) and Compact Letter Display (CLD) for summarizing results.

Related

  • Project
    statsmodels/statsmodels

    A Python package for statistical computation that provides tools for estimation and inference for a wide range of statistical models.

  • Project
    arviz-devs/arviz

    A Python package for the exploratory analysis of Bayesian models, providing tools for posterior analysis, model checking, and diagnostics.

  • Project
    numpy/numpy

    NumPy is the fundamental package for scientific computing with Python, providing N-dimensional array objects and tools for linear algebra and Fourier transforms.

  • Project
    pandas-dev/pandas

    A powerful Python data analysis toolkit that provides fast, flexible, and expressive data structures for working with relational or labeled data.

  • Project
    GAA-UAM/scikit-fda

    A Python package for Functional Data Analysis (FDA) that provides tools for the representation, preprocessing, and statistical analysis of data depending on a continuous parameter.