|
|
 |
| |
|
|
|
Asymptotic Inference for Periodic Time-Varying Bivariate Poisson INGARCH(1,1) Processes |
|
|
|
PP: 77-82 |
|
|
|
doi:10.18576/jsapl/100106
|
|
|
|
Author(s) |
|
|
|
Ahmed Ghezal,
|
|
|
|
Abstract |
|
|
| In this paper, we would like to propose an extension of bivariate Poisson integer valued GARCH (shortly, BINGARCH) processes to periodically time-varying coefficients one. In these models, the parameters are allowed to switch periodically between different seasons. The main motivation of this new model is capable of modeling bivariate time series of counts. So, a necessary and sufficient condition for the periodically stationary in the mean, is established, while providing the closed-form expression for the mean. Furthermore, we show that the conditional maximum likelihood estimator (CMLE) of the parameter of the model is strongly consistent and asymptotically normal.
|
|
|
|
|
|
|
|
