Integer Autoregressive modeling: a new gretl routine

Lucio Palazzo
(2019) In: Proceedings of the International Conference on the GNU Regression, Econometrics and Time Series Library. Naples, Italy. 13-14 June 2019. pp. 95-103, fedOA

 RG DOI Online  pdf

This paper focuses on a family of observation-driven models for autoregressive discrete-valued data, called INAR models. The main purpose of the project is to write and document a set of Gretl Econometric Software functions that perform time series estimation of one-lagged univariate INAR models with Poisson and Negative Binomial marginals.