Title: Integer-valued autoregressive models with dynamic coefficient driven by a stochastic recurrence equation
Authors: Paolo Gorgi - University of Padua (Italy) [presenting]
Abstract: A new class of integer-valued autoregressive (INAR) models with dynamic coefficient is proposed. The peculiarity of this class of models lies in the specification of the INAR coefficient through a stochastic recurrence equation. The estimation of the model can be performed by maximum likelihood and the consistency of the estimator is proved. The flexibility of the proposed specification is illustrated in a simulation study. An application to a time series of crime reports is presented. The results show how the dynamic coefficient can allow to enhance both the in-sample and the out-of-sample performance of INAR models.