Zero-inflated finite-range integer-valued time series model with dependent thinning operator

This is the authors' abstract. We don't add key points for this paper.

To model the finite-range counting time series data with dependent structure and zero inflation, we develop an extended binomial autoregressive model with dependent thinning operator, which has a bounded support and generates zeros as a negative innovational outlier. Moreover, some basic probabilistic and statistical properties of the model are derived. The unknown parameters of the model are estimated through the conditional maximum likelihood method. The asymptotic properties of such estimators are derived. In addition, this paper also studies how to test for zero inflation in the data and prediction problems. Some simulation results of the estimators and a real data analysis are presented to show the performance of the proposed model at the end of the paper.