Bayesian model selection and parameter estimation in penalized regression model using SMC samplers
| Autor: | Thi Le Thu Nguyen, François Septier, Gareth Peters, Yves Delignon |
|---|---|
| Přispěvatelé: | LAGIS-SI, Laboratoire d'Automatique, Génie Informatique et Signal (LAGIS), Université de Lille, Sciences et Technologies-Centrale Lille-Centre National de la Recherche Scientifique (CNRS)-Université de Lille, Sciences et Technologies-Centrale Lille-Centre National de la Recherche Scientifique (CNRS), Department of Statistical Science, University College of London, University College of London [London] (UCL), Septier, François |
| Předmět: | |
| Zdroj: | Heriot-Watt University 21st European Signal Processing Conference (EUSIPCO) 21st European Signal Processing Conference (EUSIPCO), Sep 2013, Marrakech, Morocco. pp.1-5 |
| Popis: | International audience; 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 coef- ficients. The performance of the proposed methodology is demonstrated in two different settings. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|