Highest Posterior Density Intervals of Unimodal Distributions As Analogues to Profile Likelihood Ratio Confidence Intervals

TL;DR

This paper demonstrates that under certain conditions, the highest posterior density interval in Bayesian analysis can serve as a transformation-invariant analogue to the frequentist profile likelihood ratio confidence interval, linking Bayesian and frequentist methods.

Contribution

The paper provides a proof establishing conditions under which the HPD interval is transformation invariant, aligning it with the properties of profile likelihood ratio confidence intervals.

Findings

HPD intervals can be transformation invariant under specific conditions

A formal proof links HPD intervals to profile likelihood ratio confidence intervals

The results bridge Bayesian and frequentist confidence interval concepts

Abstract

In Bayesian statistics, the highest posterior density (HPD) interval is often used to describe properties of a posterior distribution. As a method for estimating confidence intervals (CIs), the HPD has two main desirable properties. Firstly, it is the shortest interval to have a specified coverage probability. Secondly, every point inside the HPD interval has a density greater than every point outside the interval. However, the HPD interval is sometimes criticized for being transformation invariant. We make the case that under certain conditions the HPD interval is a natural analog to the frequentist profile likelihood ratio confidence interval (LRCI). Our main result is to derive a proof showing that under specified conditions, the HPD interval with respect to the density mode is transformation invariant for monotonic functions in a manner which is similar to a profile LRCI.