|
Activity Number:
|
276
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #307172 |
|
Title:
|
High-Dimensional Modeling and Model Selection
|
|
Author(s):
|
Carlos Carvalho*+
|
|
Companies:
|
Duke University
|
|
Address:
|
, Durham, NC, 27708-0251,
|
|
Keywords:
|
sparse factor models ; evolutionary search
|
|
Abstract:
|
We describe latent factor models for multivariate analysis in very high dimensions and classes of models that couple this framework with factor regressions for predictive modeling of multivariate response variables. We use sparse factor models---relationships between high-dimensional variables and underlying, lower-dimensional latent factors are sparse---created using sparsity-inducing priors. Model search and fitting are addressed through stochastic simulation (MCMC) and a novel evolutionary search. The latter computational approach explores, defines, and fits models for higher-dimensional problems through an evolutionary process that gradually expands the dimension of the sample space. Examples are drawn from studies in breast cancer genomics, where the sparse factor models represent observed relationships in gene expression of thousands of genes.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
