Change Point in Variance of Long-Range Dependent Processes

• Activity Number: 113 - Statistical Computing in Modern Statistics
• Type: Contributed
• Date/Time: Monday, August 8, 2022 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Computing
• Author(s): Kyungduk Ko*
• Companies: Boise State University
• Keywords: Change point; Effective Number of Independent Samples; Levene's F; Long-Range dependence

Abstract

We propose a method to detect a change point in a sequence of data with long-range dependence. As a test statistic to detect a change point in variance, a Levene-type F statistic is adopted and modified to address long-range dependent autocorrelations among data. A main modification to the statistic is done by applying effective number of independent samples (ENIS) to the test statistic, which is for removing or reducing the effect of long-range dependent autocorrelations existing in data. A simulation study is presented.