A discrete valued time series model; 


Model descriptionAs an example of a discrete valued time series we use the 'polio data' considered by Kuk & Cheng (1999). It is assumed that y_{i} has a Poisson (l_{i}) distribution, wherelog(l_{i}) = X_{i}b + u_{i}.
Here, X_{i} is a covariate vector, b is a vector of regression parameters and u_{i} is a first order autoregressive process u_{i} = r*u_{i1}
+ e_{i}.
Further details about the model can be found here: polio.pdf. 