Online Program

Return to main conference page
Friday, October 19
Fri, Oct 19, 2:30 PM - 3:45 PM
Caprice 3-4
Speed Session 4

Empirical Likelihood for Change Point Detection in Autoregressive Models (304763)

*Ramadha Piyadi Gamage, Western Washington University 
Wei Ning, Bowling Green State University 

Keywords: Autoregressive model; Change point analysis, Empirical Likelihood, Extreme value distribution, Consistency.

Testing change points has been an important issue among statisticians due to its broad applications in many practical fields such as economy, biology, quality control, etc. A change point is an observational time point at which a structural pattern change occurs during an experimentation process. Several research work has been carried out to detect if there are any changes in the sequence of observations and to estimate the corresponding locations in time series data. In this paper, a nonparametric method based on the empirical likelihood is proposed to detect the structural changes of the parameters in autoregressive (AR) models. Empirical Likelihood (EL) method combines the reliability from nonparametric methods and flexibility of parametric methods. Under certain conditions, the asymptotic null distribution of the empirical likelihood ratio test statistic is proved to be the extreme value distribution. Further, the consistency of the test statistic has been proved. Simulations have been carried out to show that the power of the proposed test statistic is significant. The proposed method is applied to real world data set to further illustrate the testing procedure.