Please use this identifier to cite or link to this item: https://doi.org/10.1198/016214507000000509
Title: Unified LASSO estimation by least squares approximation
Authors: Wang, H.
Leng, C. 
Keywords: Adaptive LASSO
Bayes information criterion
LASSO
Least angle regression
Least squares approximation
Microarray data
Oracle property
Solution path
Issue Date: Sep-2007
Citation: Wang, H., Leng, C. (2007-09). Unified LASSO estimation by least squares approximation. Journal of the American Statistical Association 102 (479) : 1039-1048. ScholarBank@NUS Repository. https://doi.org/10.1198/016214507000000509
Abstract: We propose a method of least squares approximation (LSA) for unified yet simple LASSO estimation. Our general theoretical framework includes ordinary least squares, generalized linear models, quantile regression, and many others as special cases. Specifically, LSA can transfer many different types of LASSO objective functions into their asymptotically equivalent least squares problems. Thereafter, the standard asymptotic theory can be established and the LARS algorithm can be applied. In particular, if the adaptive LASSO penalty and a Bayes information criterion-type tuning parameter selector are used, the resulting LSA estimator can be as efficient as the oracle. Extensive numerical studies confirm our theory. © 2007 American Statistical Association.
Source Title: Journal of the American Statistical Association
URI: http://scholarbank.nus.edu.sg/handle/10635/105451
ISSN: 01621459
DOI: 10.1198/016214507000000509
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

159
checked on Dec 17, 2018

WEB OF SCIENCETM
Citations

145
checked on Dec 17, 2018

Page view(s)

110
checked on Nov 2, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.