Bayesian Rank Based Methods for Linear Models
Faculty Advisor
Dr. Magdalena Niewiadomska-Bugaj
Department
Statistics
Presentation Date
4-24-2015
Document Type
Poster
Abstract
A Bayesian Rank Based Method for linear models where estimation of the regression coefficient(s) is based on the full conditional distributions is proposed in this research. The data likelihood is based on the asymptotic distribution of the gradient function where the asymptotic linearity of this rank based quantity is utilized. Prior distributions are put on regression coefficient(s) and scale parameter. The effects of different priors on this scale parameter are studied. The Markov Chain Monte Carlo method is used to simulate data from the respective full conditionals for inference purposes. The proposed method is compared to Zhan and Hettmansperger (2009) results.
WMU ScholarWorks Citation
Dzikunu, James, "Bayesian Rank Based Methods for Linear Models" (2015). Research and Creative Activities Poster Day. 157.
https://scholarworks.wmich.edu/grad_research_posters/157
Comments
This poster was presented at the 2015 Western Michigan University Research and Creative Activities Poster Day. The poster and abstract are currently unavailable through ScholarWorks.