Bayesian Prediction of canadian lynx data using FRAR model

ARMA models are used as exploratory tools for identifying, estimating and forecasting for the Canadian lynx data, which have attained benchmark status in the time series literature since the work of Moran in 1953. This paper shows that the Full Range Autoregressive (FRAR) model, free from order determination processes, provides an acceptable alternative to the more widely adopted class of ARMA models.