måndag 7 december 2015

How Frank Wilcoxon helped statisticians walk the non-parametric path

From Significance 
Part of the job of a statistician is to make inferences about a population from a sample. Sometimes we might assume that our data come from a particular distribution. In the case of the height of adult females, observations will be well fitted by a normal distribution, meaning that there are few very small and very tall people and a lot of people around the middle. In this case the distribution will be completely specified by assigning values to its parameters (here, the mean and variance). Our inferences are thus termed parametric.