• There are currently no refbacks.

Copyright (c) 2019 Pakistan Journal of Statistics and Operation Research

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.


Gibbs Sampling for Bayesian Prediction of SARMA Processes


Multiplicative SARMA models; Posterior analysis; Predictive analysis; MCMC methods; Gibbs sampler.


In this article we present a Bayesian prediction of multiplicative seasonal autoregressive moving average (SARMA) processes using the Gibbs sampling algorithm. First, we estimate the unobserved errors using the nonlinear least squares (NLS) method to approximate the likelihood function. Second, we employ conjugate priors on the model parameters and initial values and assume the model errors are normally distributed to derive the conditional posterior and predictive distributions. In particular, we show that the conditional posterior distribution of the model parameters and the variance are multivariate normal and inverse gamma respectively, and the conditional predictive distribution of the future observations is a multivariate normal. Finally, we use these closed-form conditional posterior and predictive distributions to apply the Gibbs sampling algorithm to approximate empirically the marginal posterior and predictive distributions, enabling us easily to carry out multiple-step ahead predictions. We evaluate our proposed Bayesian method using simulation study and real-world time series datasets.





Pakistan Journal of Statistics and Operation Research; Vol. 15 No. 2, 2019

Print ISSN: 1816-2711 | Electronic ISSN: 2220-5810