On robust estimation of negative binomial INARCH models

Abstract We discuss robust estimation of INARCH models for count time series, where each observation conditionally on its past follows a negative binomial distribution with constant scale parameter, and the conditional mean depends linearly previous observations. develop several estimators, some them being computationally fast modifications methods moments, rather efficient maximum likelihood. These estimators are compared to related recent proposals using simulations. The usefulness proposed is illustrated by real data example.