We compare sparse and dense representations of predictive models in macroeconomics, microeconomics, finance. To deal with a large number possible predictors, we specify prior that allows for both variable selection shrinkage. The posterior distribution does not typically concentrate on single model, but wide set often include many predictors.

In this article, we are interested in detecting latent group structures and significant covariates a high-dimensional panel data model with both individual time fixed effects. The slope coefficients of the assumed to be subject dependent, there exist where homogeneous within groups heterogeneous between groups. We develop penalized estimator for recovering sparsity patterns simultaneously. prop...