Bell-INGARCH Model (2311.11352v1)

Abstract: Integer-valued time series exist widely in economics, finance, biology, computer science, medicine, insurance, and many other fields. In recent years, many types of models have been proposed to model integer-valued time series data, in which the integer autoregressive model and integer-valued GARCH model are the most representative. Although there have been many results of integer-valued time series data, the parameters of integer-valued time series model structure are more complicated. This paper is dedicated to proposing a new simple integer-valued GARCH model. First, the Bell integer-valued GARCH model is given based on Bell distribution. Then, the conditional maximum likelihood estimation method is used to obtain the estimators of parameters. Later, numerical simulations confirm the finite sample properties of the estimation of unknown parameters. Finally, the model is applied in the two real examples. Compared with the existing models, the proposed model is more simple and applicable.