Bayesian estimation supersedes the t-test


License
MIT
Install
pip install best==2.0.5

Documentation

BEST: Bayesian Estimation Supersedes the t-Test

Python implementation of a Bayesian model to replace t-tests with Bayesian estimation, following the idea described in the following publication:

John K. Kruschke. Bayesian estimation supersedes the t test. Journal of Experimental Psychology: General, 2013, v.142 (2), pp. 573-603. (doi: 10.1037/a0029146)

The package implements Bayesian estimation of the mean of one or two groups, and plotting functions for the posterior distributions of variables such as the effect size, group means and their difference.

Documentation

See the documentation for more information, at best.readthedocs.io.

Requirements

Examples

A complete analysis and plotting is done in just a few lines:

>>> best_out = best.analyze_two(group1_data, group2_data)
>>> fig = best.plot_all(best_out)
>>> fig.savefig('best_plots.pdf')

For example, the two-group analysis in examples/smart_drug.py produces the following output:

smart_drug.png

More detailed analysis of the same data can be found in the Jupyter notebook examples/Smart drug (comparison of two groups).ipynb.

An example single-group analysis can be found in examples/paired_samples.py.

The documentation describes the API in detail.

Installation

Ensure your Python version is sufficiently up-to-date (at least 3.7):

$ python --version
Python 3.7.12

Then install this package and its dependencies with PIP:

$ pip install best

Developer notes

Tests

Running the tests requires pytest (pip install pytest), and then can be run with:

$ python -m pytest tests

The plotting tests only ensure that the plot_all function does not throw errors, and the plots need manual verification at tests/data/plot_all_*.pdf.

Documentation

The documentation can be built with Sphinx:

$ cd docs
$ make html