Zero-inflated count time series models using Gaussian copula

Abstract

Count time series data are observed in several applied disciplines such as environmental science, biostatistics, economics, public health, and finance. In some cases, a specific count, usually zero, may occur more often than other counts. However, overlooking the frequent occurrence of zeros could result in misleading inferences. In this article, we develop a copula-based time series regression model for zero-inflated counts. Zero-inflated Poisson (ZIP), zero-inflated negative binomial (ZINB), and zero-inflated Conway-Maxwell-Poisson (ZICMP) distributed marginals will be considered, and the joint distribution is modeled under Gaussian copula with autoregression moving average (ARMA) errors. Sequential sampling likelihood inference is performed. Simulated and real-life data examples are provided and studied to evaluate the proposed method.