Abstract There are several robust regression methods that can be used to select parsimonious models in regression. In this paper, a study of the quantile regression with a normal-compound gamma scale mixture prior is presented. The Monte Carlo Markov Chain (MCMC) is derived for posterior inference. Finally, the robustness of this model is demonstrated using both real and simulated data. The results show that the proposed method performs very well compared to some of the existing methods.
Field : Mühendislik
Journal Type : Uluslararası
Relevant Articles | Author | # |
---|
Article | Author | # |
---|