Univ. Heidelberg
Statistics Group   Institute for Mathematics   Faculty of Mathematics and Computer Science   University Heidelberg
Ruprecht-Karls-Universität Heidelberg Institute for Mathematics Statistics of inverse problems Research Group
german english french



Publications
Cooperations
Research projects
Events
Teaching
Completed theses
People
Contact


Last edited on
Oct 17, 2024 by JJ
.
Thesis:
Bachelor in Mathematics

Author:
Kai Becker

Title:
Der adaptive Lasso-Schätzer

Supervisors:
Stefan Richter (Universität Heidelberg)
Jan JOHANNES

Abstract:
Since variabel selection by the lasso estimator is inconsistent under certain conditions, we introduce a weighted lasso estimator, called adaptive lasso estimator. In this approach adaptive wheights are introduced, to penalize the various coefficients in the l1 penalty term. Subsequently, we proof that the adaptive Lasso estimator satisfies the oracle properties. This implies that, firstly, it provides asymptotically optimal estimation, and, secondly, it maintains consistency in variable selection. Furthermore, we compare the estimation achieved using the adaptive lasso estimator with that of the Lasso estimator in several simulations.

References:
Hui Zou. The adaptive Lasso and its oracle properties, Journal of the American Statistical Association, 101(476):1418-1429, 2006.