Robust High-Dimensional Testing Using a Minimum Hellinger Distance Procedure

• Activity Number: 161
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Bret Hanlon*+
• Keywords: minimum hellinger distance ; robust ; infinite dimensional framework

Abstract

Technical advances in biomedicine have produced an abundance of high throughput data and have stimulated substantial interest in high-dimensional two-sample problems. Recently, Kuelbs and Vidyashankar (2010, Annals of Statistics) developed an infinite-dimensional framework to study the comparisons of means when the number of parameters increase with the sample size. They demonstrate via simulations and asymptotic theory that their methods have accurate Type I error and high power. In this talk we describe an extension of their work to allow for robust testing utilizing a minimum Hellinger distance procedure. We illustrate our results using both simulated and experimental data.