The effect size is a common way to describe a difference between two distributions. When these distributions are normal, one of the most popular approaches to express the effect size is Cohen's d. Unfortunately, it doesn't work great for non-normal distributions.
In this post, I will show a robust Cohen's d-consistent effect size formula for nonparametric distributions.Read more
Outlier detection is an important step in data processing. Unfortunately, if the distribution is not normal (e.g., right-skewed and heavy-tailed), it's hard to choose a robust outlier detection algorithm that will not be affected by tricky distribution properties. During the last several years, I tried many different approaches, but I was not satisfied with their results. Finally, I found an algorithm to which I have (almost) no complaints. It's based on the double median absolute deviation and the Harrell-Davis quantile estimator. In this post, I will show how it works and why it's better than some other approaches.Read more