Objective Bayesian Two Sample Hypothesis Testing for Online Controlled Experiments

by Alex Deng

WWW 2015 1st Workshop of Online and Offline Evaluation of Web-based Services,

May, 2015, Florence, Italy. Paper, Slides(PDF). 

Abstract

As A/B testing gains wider adoption in the industry, more people begin to realize the limitations of the traditional frequentist null hypothesis statistical testing (NHST). The large number of search results for the query “Bayesian A/B testing” shows just how much the interest in the Bayesian perspective is growing. In recent years there are also voices arguing that Bayesian A/B testing should replace frequentist NHST and is strictly superior in all aspects. Our goal here is to clarify the myth by looking at both advantages and issues of Bayesian methods. In particular, we propose an objective Bayesian A/B testing framework for which we hope to bring the best from Bayesian and frequentist methods together. Unlike traditional methods, this method requires the existence of historical A/B test data to objectively learn a prior. We have successfully applied this method to Bing, using thousands of experiments to establish the priors.