/resample

Randomization-based inference in Python

Primary LanguagePythonBSD 3-Clause "New" or "Revised" LicenseBSD-3-Clause

resample

Description

resample provides a set of tools for performing randomization-based inference in Python, primarily through the use of bootstrapping methods and Monte Carlo permutation tests. Documentation can be found on Read the Docs.

Features

  • Bootstrap samples (ordinary or balanced, both with optional stratification) of arrays with arbitrary dimension
  • Parametric bootstrap samples (Gaussian, Poisson, gamma, etc.) of one-dimensional arrays
  • Bootstrap confidence intervals (percentile or BCa) for any well-defined parameter
  • Jackknife estimates of bias and variance
  • Randomization-based variants of traditional statistical tests (t-test, ANOVA F-test, K-S test, etc.)
  • Tools for working with empirical distributions (cumulative distribution, quantile, and influence functions)

Dependencies

Installation requires numpy and scipy.

Installation

The latest release can be installed from PyPI:

pip install resample

or using conda:

conda install resample -c conda-forge