Estimating Change Point by a Weighted CUSUM Type Statistic

10.29973/JCSA.200506.0004

Hong-Ling Zhou；Kung-Yee Liang

Change point ； CUSUM ； bootstrap confidence interval

中國統計學報

43卷2期（2005 / 06 / 01）

183 - 198

英文

In this paper we consider the change point model with single abrupt change in a sequence of independent random variables. An estimation procedure for the change point based on a weighted CUSUM type statistic is proposed. A bootstrap resampling scheme is applied to obtain the confidence interval of the proposed estimator. Comparisons are made between the proposed procedure and the maximum likelihood estimation in terms of precision of the estimates.

1. Boukai, B.(1993).A nonparametric bootstrapped estimate of the change-point.Journal of Nonparametric Statistics,123-134.
2. Davison, A. C.,Hinkley, D. V.(1997).Bootstrap Methods and their Application.Cambridge University Press.
3. Hinkley, D. V.(1971).Inference about the change-point from cumulative sum test.Biometrika,58,509-523.
4. Hinkley, D. V.,Hinkley, E. A.(1970).Inference about the change-point in a sequence of binomial variables.Biometrika,57,477-488.
5. Page, E. S.(1954).Continuous inspection schemes.Biometrika,41,100-114.
6. Pettitt, A. N.(1980).A simple cumulative sum type statistic for the change-point problem with zero-one observations.Biometrika,67,79-84.
7. Worsley, K. J.(1983).The power of likelihood ratio and cumulative sum tests for a change in a binomial probability.Biometrika,70,455-464.
8. Zacks, S.(1983).Survey of classical and bayesian approaches to the change-point problem: Fixed sample and sequential procedures of testing and estimation.Recent Advances in Statistics,245-269.