Abstract
A Bayesian approach to modeling a rich class of nonconjugate problems is presented. An adaptive Monte Carlo integration technique known as the Gibbs sampler is proposed as a mechanism for implementing a conceptually and computationally simple solution in such a framework. The result is a general strategy for obtaining marginal posterior densities under changing specification of the model error densities and related prior densities. We illustrate the approach in a nonlinear regression setting, comparing the merits of three candidate error distributions.
Original language | English (US) |
---|---|
Pages (from-to) | 399-405 |
Number of pages | 7 |
Journal | Canadian Journal of Statistics |
Volume | 19 |
Issue number | 4 |
DOIs | |
State | Published - Dec 1991 |
Keywords
- Bayesian model choice
- Gibbs sampler
- nonlinear models
- nonnormal errors