This paper is based on a plenary presentation at NORDSTAT Meeting in Voss Norway, June 2010.
1467-9469/asset/olbannerleft.jpg?v=1&s=7c6e3695f9a27dcc0a4d95807a2fe24ce7ae470c)
1467-9469/asset/olbannercenter.png?v=1&s=04f72f4a46fa7ce2f5772acab5158df122071db6)
1467-9469/asset/olbannerright.jpg?v=1&s=d092c0c4f28b019885c4c47a643aa339df452b86)
Estimation for High-Dimensional Linear Mixed-Effects Models Using ℓ1-Penalization
Article first published online: 10 MAY 2011
DOI: 10.1111/j.1467-9469.2011.00740.x
© 2011 Board of the Foundation of the Scandinavian Journal of Statistics
Additional Information
How to Cite
SCHELLDORFER, J., BÜHLMANN, P. and DE GEER, S. V. (2011), Estimation for High-Dimensional Linear Mixed-Effects Models Using ℓ1-Penalization. Scandinavian Journal of Statistics, 38: 197–214. doi: 10.1111/j.1467-9469.2011.00740.x
Publication History
- Issue published online: 10 MAY 2011
- Article first published online: 10 MAY 2011
- Received October 2010, in final form February 2011
Keywords:
- adaptive Lasso;
- coordinate gradient descent;
- coordinatewise optimization;
- Lasso;
- random-effects model;
- variable selection;
- variance components
Abstract. We propose an ℓ1-penalized estimation procedure for high-dimensional linear mixed-effects models. The models are useful whenever there is a grouping structure among high-dimensional observations, that is, for clustered data. We prove a consistency and an oracle optimality result and we develop an algorithm with provable numerical convergence. Furthermore, we demonstrate the performance of the method on simulated and a real high-dimensional data set.