Nonhomogeneous Poisson Process Inference When There Are Missing Counts

• Activity Number: 694
• Type: Contributed
• Date/Time: Thursday, August 13, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Peng Liu* and Rajneesh Rajneesh
• Companies: SAS Institute and SAS Institute
• Keywords: reliability ; NHPP ; growth model ; repairable system ; missing information

Abstract

Recurrence data are usually recorded in two forms: the time to the recurrences, or the number of recurrences in disjoint time intervals. A data can also be a mixture of both forms. A non-homogeneous Poisson process (NHPP) model can analyze some of those data forms. We will discuss here a situation where the numbers of recurrences are missing during one or more time periods. Crow and Basu [1, 2] studied the problem and proposed an approach under the name of Gap Analysis. In this paper, we propose our approach and report our simulation study that compares our method to the Gap Analysis.