الفهرس | Only 14 pages are availabe for public view |
Abstract The study develops a Bayesian forecasting for the multiplicative Double Seasonal Autoregressive models. Two different approaches are employed; the first is an approximate Bayesian forecasting and the second one is Gibbs sampling approach. A normal inverse gamma prior is used because it is a conjugate class. The adequacy of both proposed Bayesian approaches is checked using four simulated examples and a real data set. Empirical results indicate the accuracy of Bayesian forecasts where these Bayesian forecasts lie within the credible intervals |