Hierarchical Bayesian Models of Change-Point Problem Using Bayes Factor

• Activity Number: 352
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Younshik Chung*+ and Seongho Song and Myoungjin Jung
• Companies: Pusan National University and University of Cincinnati and Pusan National University
• Keywords: Hierarchical bayesian Model ; Change Point problem ; Bayes Factor ; Skew Normal distribution

Abstract

In this paper, we consider the Hierarchical Bayesian models of change-point problem in a sequence of random variables having either normal population or skew-normal population. Further, we consider the problem of detecting an influential point concerning the change-point using Bayes factor. Our proposed models are illustrated with the real example, the annual flow volume data of Nile river at Aswan from 1871 to 1970. The result studied by Jong and Penzer (1998) indicated only strong candidated for outliers in the years 1877 and 1913. However, the result using our proposed models indicated the largest influential observation in the year 1888 among outliers. We have shown that K_d is useful to measure the influence of observations on Bayes factors. We have considered omitting single observations. The result of K_d studied by Pettit and Young (1990) indicated also the strong candidated for outliers in the years 1877 and 1913.