|
Activity Number:
|
100
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #305026 |
|
Title:
|
Bayesian Inference for the Stereotype Regression Model: Application to a Case-Control Study of Prostate Cancer
|
|
Author(s):
|
Jaeil Ahn*+ and Bhramar Mukherjee
|
|
Companies:
|
University of Michigan and University of Michigan
|
|
Address:
|
Biostatistics, 1420 Washington Heights, Ann Arbor, MI, 48105,
|
|
Keywords:
|
stereotype regression ; log-odds ; baseline category model ; case-control study ; bayesian
|
|
Abstract:
|
The stereotype regression model for categorical outcomes, proposed by Anderson is nested between the baseline category logits and adjacent category logits model with proportional odds structure. The stereotype model is more parsimonious than the ordinary baseline-category model due to a product representation of the log odds-ratios in terms of a common parameter for each predictor and category specific scores. The advantage of the stereotype model is that it provides unbiased and valid inference under outcome-stratified sampling as in case-case-controls studies. We present comprehensive Bayesian inference and model comparison procedure for this class of models which circumvents many of the limitations of the classical frequentist approach.
|
