Machine Learning with R Cookbook - Second Edition

How it works...

In this recipe, we discuss a nonparametric test method, the Wilcoxon Rank Sum test (also known as the Mann-Whitney U-test). For a student's t-test, it is assumed that the differences between the two samples are normally distributed (and it also works best when the two samples are normally distributed). However, when the normality assumption is uncertain, one can adopt the Wilcoxon Rank Sum Test to test a hypothesis.

Here, we used a Wilcoxon Rank Sum test to determine whether the mpg of automatic and manual transmission cars in the dataset, mtcars, is distributed identically. From the test result, we see that the p-value = 0.001871 (< 0.05) rejects the null hypothesis, and also reveals that the distribution of mpg in automatic ...