Griffin, Jim E. and Brown, Philip J. (2007) Bayesian adaptive lassos with non-convex penalization. Working Paper. Coventry: University of Warwick. Centre for Research in Statistical Methodology. Working papers, Vol.2007 (No.2).

Preview

PDF WRAP_Griffin_07-2wv2.pdf - Published Version - Requires a PDF viewer. Download (627Kb)

Request Changes to record.

Abstract

The lasso (Tibshirani,1996) has sparked interest in the use of penalization of
the log-likelihood for variable selection, as well as shrinkage. Recently, there have
been attempts to propose penalty functions which improve upon the Lassos properties
for variable selection and prediction, such as SCAD (Fan and Li, 2001) and
the Adaptive Lasso (Zou, 2006). We adopt the Bayesian interpretation of the Lasso
as the maximum a posteriori (MAP) estimate of the regression coefficients, which
have been given independent, double exponential prior distributions. Generalizing
this prior provides a family of adaptive lasso penalty functions, which includes the quasi-cauchy distribution (Johnstone and Silverman, 2005) as a special case.
The properties of this approach are explored. We are particularly interested in the
more variables than observations case of characteristic importance for data arising
in chemometrics, genomics and proteomics - to name but three. Our methodology
can give rise to multiple modes of the posterior distribution and we show how
this may occur even with the convex lasso. These multiple modes do no more
than reflect the indeterminacy of the model. We give fast algorithms and suggest
a strategy of using a set of perfectly fitting random starting values to explore
different regions of the parameter space with substantial posterior support. Simulations
show that our procedure provides significant improvements on a range of
established procedures and we provide an example from chemometrics.

Item Type:	Working or Discussion Paper (Working Paper)
Subjects:	Q Science > QA Mathematics
Divisions:	Faculty of Science > Statistics
Library of Congress Subject Headings (LCSH):	Regression analysis
Series Name:	Working papers
Publisher:	University of Warwick. Centre for Research in Statistical Methodology
Place of Publication:	Coventry
Official Date:	2007
Dates:	Date Event 2007 Published
Date of first compliant deposit:	1 August 2016
Volume:	Vol.2007
Number:	No.2
Number of Pages:	30
Status:	Not Peer Reviewed
Access rights to Published version:	Open Access

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads