In this paper, we consider the structural change in a class of discrete valued time series, where true conditional distribution observations is assumed to be unknown. The mean process depends on parameter θ∗ which may over time. We provide sufficient conditions for consistency and asymptotic normality Poisson quasi-maximum likelihood estimator (QMLE) model. an epidemic change-point detection pr...

Change point detection in time series has attracted substantial interest, but most of the existing results have been focused on detecting change points domain. This paper considers situation where nonlinear potential state We apply a density-weighted anti-symmetric kernel function to domain and therefore propose nonparametric procedure test existence points. When is affirmative, we further intr...