A Geometric Integer – Valued GARCH Model

Abstract

AbstractIn this paper, the modeling of integer – valued time series with over dispersion is discussed. A particularly flexible model for time series of counts is the Geometric integer – valued Generalized Autoregressive conditional hetrosscedatic (GEO - GARCH) model which properly accounts for the over disersion and non – negativity. This model is proposed and the autocorrelation function are given. The main goal of this paper is to study the properties of model and some results from simulation study to estimate the unknown parameters by using Maximum Likelihood method.