A New Approach to Bayesian Change Point Detection Using Lifting Wavelet Transform

Authors

Arunendu Chatterjee, Department of Mathematics, University of Wisconsin-River Falls, 410 S. 3rd Street, River Falls, WI 54022, USAFollow

Author Country (or Countries)

USA

Abstract

In this paper, we use wavelets in a Bayesian context to identify changes in the pattern of data collected over time in the presence of noise and missing observations in the data. A Bayesian analysis based on the wavelet coefficients applying lifting is discussed to identify change points. Based on a simulation study, recommendations are made on the choice of lifting wavelet coefficients in the presence of noise and missing observations using an adaptive lifting technique. We apply our algorithm to a real data problem where change points are already known to illustrate our recommendations

Suggested Reviewers

N/A

Digital Object Identifier (DOI)

http://dx.doi.org/10.18576/jsap/060301

Recommended Citation

Chatterjee, Arunendu (2017) "A New Approach to Bayesian Change Point Detection Using Lifting Wavelet Transform," Journal of Statistics Applications & Probability: Vol. 6: Iss. 3, Article 1.
DOI: http://dx.doi.org/10.18576/jsap/060301
Available at: https://digitalcommons.aaru.edu.jo/jsap/vol6/iss3/1