Comparing the performances of the priors on the Bayesian estimation of change point location in a sequence of normal random variables
Abstract
In this paper, the problem of one change point occurring simultaneously in both the mean and variance of a sequence of normal random variables is considered. The Bayesian method is used for the estimation of change point location; and the marginal posterior distributions of the change point location are obtained under the assumptions of noninformative and informative prior distributions for the parameters. The performances of the prior distributions on the estimation of the change point location, with different sample sizes, are investigated via extensive simulation studies.
Keywords
References
- [1] J.M. Bernardo, Noninformative priors do not exist: a discussion with José M. Bernardo, J. Statist. Planning and Inference 65 (1997), pp. 159-189 (to appear, with discussion)
- [2] J.M. Bernardo, E. Gutierrcz-Pefia and A.F.M. Smith, Comment to 'Exponential and Bayesian conjugate families: review and extensions', Test 6 (1997), pp. 70-71.
- [3] H. Boudjellaba, B. MacGibbon, P. Sawyer, On exact inference for change in a Poisson Sequence, Communications in Statistics A: Theory and Methods 30(3) (2001), pp. 407434.
- [4] J. Chen, A.K. Gupta, Testing and locating variance change points with application to stock prices, Journal of the American Statistical Association: JASA 92(438) (1997), pp.739-747.
- [5] J. Chen, A.K. Gupta, Change point analysis of a Gaussian mode, Statistical Papers 40 (1999), pp. 323–333.
- [6] J. Chen, A.K. Gupta, On change point detection and estimation, Communications in Statistics-Simulation and Computation 30 (2001), pp. 665-697.
- [7] J. Chen, A.K. Gupta, Statistical inference of covariance change points in Gaussian model, Statistics 38 (2004), pp. 17-28.
- [8] J. Chen, A.K. Gupta, A Bayesian approach to the statistical analysis of a smooth–abrupt change point analysis, Advances and Applications in Statistics 7 (2007), pp. 115-125.
Details
Primary Language
English
Subjects
-
Journal Section
Research Article
Authors
Ayten Yiğiter
*
0000-0001-8180-995X
Türkiye
Publication Date
December 30, 2017
Submission Date
July 26, 2017
Acceptance Date
November 13, 2017
Published in Issue
Year 2017 Volume: 10 Number: 2