Asymptotics of non-linear lasso type estimators

This paper extends the linear lasso estimators to non-linear case. We are especially interested in GMM type of Lasso estimators. Lasso estimators are generalization of ridge regression in least squares. With this setup we can deal with identification problem introduced by weak instruments. When the instruments are weak, we obtain asymptotically normal limits although the rate of convergence of the estimators are slower than square root

MoreLess

Year of publication:	2004-08-11
Authors:	caner, mehmet
Institutions:	Econometric Society
Subject:	ridge regreesion \| empirical process theory \| weak instruments

More details

Series:	Econometric Society 2004 North American Winter Meetings.
Type of publication:	Book / Working Paper
Notes:	The text is part of a series Econometric Society North American Winter Meetings 2004 Number 156
Classification:	C51 - Model Construction and Estimation
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005328987