Gibbs Sampling in AR Models with Random Walk Priors
The paper analyses univariate autoregressive AR(p) models with tightness prior. The framework of the model is the conjugate normal linear model where the prior distribution is assumed to be a random walk process. The deviation from the prior distribution is measured by the tightness (hyper-) parameter λ. It is shown how the estimation of the starting values can be incorporated into the Gibbs sampling scheme. We demonstrate this approach with simulated and economic time series. It is found that for typical economic sample size the sampling fluctuations influence the posterior distribution considerably and informative prior distributions seem to be useful, especially for prediction.
KeywordsPosterior Distribution Prior Distribution Gibbs Sampler Random Walk Process Full Conditional Distribution
Unable to display preview. Download preview PDF.
- Marriott J., Ravishanker N., Gelfand A.E., Pai J. (1992): Bayesian analysis of ARMA processes: Complete sampling based inference under full likelihood, mimeo, University of Connecticut.Google Scholar
- Polasek W. (1993): Gibbs sampling in VAR models with tightness priors, mimeo, University of Basel.Google Scholar
- Polasek W. (1994): Gibbs sampling in B-VAR models with latent variables. WWZ-discussion papers Nr. 9415, University of Basel.Google Scholar