The effects of three different priors for variance parameters in the normal-mean hierarchical model

Many prior distributions are suggested for variance parameters in the hierarchical model. The “Non-informative” interval of the conjugate inverse-gamma prior might cause problems. I consider three priors – conjugate inverse-gamma, log-normal and truncated normal for the variance parameters and do the numerical analysis on Gelman’s 8-schools data. Then with the posterior draws, I compare the Bayesian credible intervals of parameters using the three priors. I use predictive distributions to do predictions and then discuss the differences of the three priors suggested.

Name thesis:degree_name

Master of Science in Statistics

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Statistics

Grantor

University of Texas at Austin

Year dc:date.issued

2010

Author dc:creator

• Chen, Zhu, 1985-

Advisor dc:contributor.advisor

• Greenberg, Betsy S.

Committee member dc:contributor.committeemember

• Sager, Thomas W.

• Bayesian inference
• Conjugate prior
• Log-normal
• Truncated normal
• Hierarchical model
• Metropolis-Hastings within Gibbs
• Gibbs sampler

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/2152/ETD-UT-2010-05-1516

OAI identifier oai:identifier

oai:repositories.lib.utexas.edu:2152/ETD-UT-2010-05-1516