Penalised Euclidean distance regression

Vasiliu, Daniel and Dey, Tanujit and Dryden, Ian L. (2017) Penalised Euclidean distance regression. Stat . ISSN 2049-1573 (In Press)

[img] PDF - Repository staff only until 13 December 2018. - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (304kB)

Abstract

A method is introduced for variable selection and prediction in linear regression problems where the number of predictors can be much larger than the number of observations. The methodology involves minimising a penalised Euclidean distance, where the penalty is the geometric mean of the $\ell_1$ and $\ell_2$ norms of the regression coefficients. This particular formulation exhibits a grouping effect, which is useful for model selection in high dimensional problems. Also, an important result is a model consistency theorem, which does not require an estimate of the noise standard deviation. An algorithm for estimation is described, which involves thresholding to obtain a sparse solution. Practical performances of variable selection and prediction are evaluated through simulation studies and the analysis of real datasets.

Item Type: Article
Schools/Departments: University of Nottingham, UK > Faculty of Science > School of Mathematical Sciences
Depositing User: Dryden, Professor Ian
Date Deposited: 13 Dec 2017 13:52
Last Modified: 19 Dec 2017 18:41
URI: http://eprints.nottingham.ac.uk/id/eprint/48710

Actions (Archive Staff Only)

Edit View Edit View