A Graph-Based Test for Comparing Multivariate Distributions in
the Multi Sample Framework
Description
We introduce a nonparametric, graphical test based on optimal matching for assessing whether multiple unknown multivariate probability distributions are equal. This method is consistent, and does not make any distributional assumptions on the data. Our procedure combines data that belong to different classes or groups to create a graph on the pooled data, and then utilizes the number of edges connecting data points from different classes to examine equality of distributions among the classes. The functions available through this package implement the work described here: .