Generalized Quasilikelihood Inference for Zero Inflated Longitudinal Count Data

Abstract

In this paper, we extend an observation-driven model for time series of zero inflated count data to longitudinal data setup. Basic properties of the models are discussed. For statistical inference of the proposed model, a generalized quasilikelihood (GQL) estimating equation has been derived for the regression parameter. A pharmaceutical data has been reanalyzed using the proposed approach and results are compared. The proposed approach produces similar estimates as given in the earlier work with much smaller standard errors.