Self-exciting threshold -valued autoregressive processes for non-stationary time series of counts

Han LiXiaoman ChenXiaogang Donga School of Mathematics and Statistics,Changchun University of Technology,Changchun,People's Republic of Chinab School of Science,Changchun University,Changchun,People's Republic of China
DOI: https://doi.org/10.1080/02331888.2024.2328594
IF: 2.346
2024-03-27
Statistics
Abstract:To handle non-stationary integer-valued time series of counts with piecewise characteristics and linear trends, this paper introduced a class of self-exciting threshold Z -valued autoregressive processes. The process is defined with a free type distribution for the innovation, which enhances the flexibility of the model. The basic probabilistic and statistical properties of the proposed model are discussed. Conditional least squares (CLS) estimator and modified quasi-likelihood (MQL) estimator, as well as their asymptotic properties are obtained. A searching algorithm for estimating the threshold parameter is also provided. Some simulation studies are conducted to show the performances of the proposed methods. Finally, an application to a real data example is provided.
statistics & probability
What problem does this paper attempt to address?