Abstract
Penalized regression methods have received a great deal of attention in recent years, mostly through frequentist models using ℓ1- regularization. However, all existing works assume that the design matrix, that links the explanatory variables to the observed response, is known a priori. Unfortunately, this is often not the case and thus solving this challenging problem is of considerable interest. In this paper, we look at a fully Bayesian formulation of this problem. This paper proposes the use of Sequential Monte Carlo samplers for joint model selection and parameter estimation. Furthermore, a new class of priors based on α-stable family distribution is proposed as non-convex penalty for regularization of the regression coefficients. The performance of the proposed methodology is demonstrated in two different settings.
Original language | English |
---|---|
Title of host publication | 21st European Signal Processing Conference (EUSIPCO 2013) |
Publisher | IEEE |
ISBN (Print) | 9780992862602 |
Publication status | Published - 8 May 2014 |
Event | 21st European Signal Processing Conference 2013 - Morocco, Marrakech, Morocco Duration: 9 Sept 2013 → 13 Sept 2013 |
Conference
Conference | 21st European Signal Processing Conference 2013 |
---|---|
Abbreviated title | EUSIPCO 2013 |
Country/Territory | Morocco |
City | Marrakech |
Period | 9/09/13 → 13/09/13 |
Keywords
- Bayesian Inference
- Model Selection
- Regularization
- SMC sampler
ASJC Scopus subject areas
- Signal Processing
- Electrical and Electronic Engineering