This repository includes an R script and JAGS model file to run the methods presented in the following publication (Curtis and Ghosh. 2011. A bayesian approach to multicolllinearity and the simultaneous selection and clustering of predictors in linear regression. Journal of Statistical Theory and Practice 5(4): 715-735.) on the Dimensions ZA data.
- Re-read paper an formulate questions for Sujit and Xiaojing
- We need to determine how the clustering of regression coefficients is
carried out. This may require contacting the authors as well.
- Also look into possibility of using phylogenetic methods, particularly majority rule consensus trees, to get clustering.
- Determine how to include categorical and factor variables.
- Advance this to a multi-response model.