Work

Generalized double Pareto shrinkage

Year: 2013

Type: article

Abstract: -like tail behavior. Bayesian computation is straightforward via a simple Gibbs sampling algorithm. We investigate the properties of the maximum a posteriori estimator, as sparse estimation plays an important role in many problems, reveal connections with some well-established regularization procedures, and show some asymptotic results. The performance of the prior is tested through simulations and an application.

Source: Statistica Sinica

DOI: 10.5705/ss.2011.048

Authors: Artin Armagan, David B. Dunson, Jaeyong Lee

Institutions: SAS Institute (United States), Duke University, Seoul National University

Language: English

Topic: Statistical Methods and Inference

Subfield: Statistics and Probability

Field: Mathematics

Domain: Physical Sciences

Open Access status: green