Integer valued AR (INAR) processes are perfectly suited for modelling count
data. We consider the inclusion of explanatory variables into the INAR
model to extend the applicability of INAR models. This greatly extends the
range of time series data sets to which INAR models can be applied and
offers an alternative to Poisson regression models. An efficient MCMC algorithm
is constructed to analyze the model and incorporates both explanatory
variable and order selection. The applicability of the methodology is demonstrated
by considering three different data sets; monthly polio incidences in
the USA 1970-1983, monthly benefit claims from the logging industry to the
British Columbia Workers’ Compensation Board 1985-1994 and the daily
score achieved by a schizophrenic patient in a test of perceptual speed.