Unified LASSO Estimation by Least Squares Approximation
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 o...
Saved in:
Published in | Journal of the American Statistical Association Vol. 102; no. 479; pp. 1039 - 1048 |
---|---|
Main Authors | , |
Format | Journal Article |
Language | English |
Published |
Alexandria
Taylor & Francis
01.09.2007
American Statistical Association Taylor & Francis Ltd |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | 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. |
---|---|
Bibliography: | ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 23 |
ISSN: | 0162-1459 1537-274X |
DOI: | 10.1198/016214507000000509 |