Uchechukwu
Department of Statistics, Michael Okpara University of Agriculture, Umudike. Abia State
Title: Bayesian statistics
Biography
Biography: Uchechukwu
Abstract
The study proposed a modified stochastic search variable selection method for selecting restriction in VAR models. This is done by eliciting a new class of modified SSVS prior using diffuse prior for the variance covariance which allows for non-diagonal treatment of the variance covariance matrix. The performance of the modified stochastic variable was evaluated using a Monte Carlo experiment and concludes by assessing the out of sample forecast performance of three macroeconomic series of Nigeria with the modified stochastic variable selection in VAR model. The modified stochastic search variable selection method was compared with existing VAR models in literature using the Mean square Error. The modified stochastic search variable selection method outperforms most of the existing bayesian and non-bayesian VAR models