Comments on "Detecting Outliers in Gamma Distribution" by M. Jabbari Nooghabi et al. (2010)

TL;DR

This paper critically examines previous methods for detecting outliers in Gamma distributions, revealing that some proposed statistical tests and their underlying assumptions are invalid in certain cases.

Contribution

It identifies flaws in existing outlier detection techniques for Gamma distributions and clarifies the conditions under which their assumptions do not hold.

Findings

Previous tests are invalid in some cases

The probability density functions used are not always valid

Outlier detection methods need revision for Gamma distributions

Abstract

This note shows that the results presented by Jabbari Nooghabi et al. (2010) do not hold in all expected cases. With this, the technique proposed by Kumar and Lalhita (2012) for detecting upper outliers in Gamma samples is also not valid. Specifically, this note shows that the probability density functions (pdf) under the null hypothesis of the test statistics therein proposed are not always valid.