The Bayesian Estimate of Vector Autoregressive Model Parameters Adopt Informative Prior Information
This research included the bayesian estimate for vector Autoregressive model with rank (p) in addition to statistical tests and predict Bayesian when the random error of model followed generalized multivariate modified Bessel distribution. The prior information about the parameters of model is represented by probability distributions belong to conjugate families. It found that the posterior marginal probability distribution for parameters matrix (Φ) is a Matrix t distribution, The posterior marginal probability distribution of covariance matrix (Σ) is uncommon and the predictive probability distribution of future observations vector is multivariate t distribution.
How to Cite
Copyright (c) 2023 Tikrit Journal of Pure Science
This work is licensed under a Creative Commons Attribution 4.0 International License.