摘要：Change point problem has been studied extensively since 1950sdue to its broad applications in many elds such as nance, biology and soon. As a special case of the multiple change point problem, the epidemicchange point problem has received a lot of attention especially in medicalstudies. In this paper, a nonparametric method based on the empiricallikelihood is proposed to detect the epidemic changes of the mean afterunknown change points. Under some mild conditions, the asymptotic nulldistribution of the empirical likelihood ratio test statistic is proved to be theextreme distribution. The consistency of the test is also proved. Simulationsindicate that the test behaves comparable to the other available tests while itenjoys less constraint on the data distribution. The method is applied to theStandford heart transplant data and detects the change points successfully.
