Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/86650 
Year of Publication: 
2006
Series/Report no.: 
Tinbergen Institute Discussion Paper No. 06-058/4
Publisher: 
Tinbergen Institute, Amsterdam and Rotterdam
Abstract: 
We propose and study a class of regression models, in which the mean function is specified parametrically as in the existing regression methods, but the residual distribution is modeled nonparametrically by a kernel estimator, without imposing any assumption on its distribution. This specification is different from the existing semiparametric regression models. The asymptotic properties of such likelihood and the maximum likelihood estimate (MLE) under this semiparametric model are studied. We show that under some regularity conditions, the MLE under this model is consistent (as compared to the possibly pseudo consistency of the parameter estimation under the existing parametric regression model), and is asymptotically normal with rate sqrt{n} and efficient. The nonparametric pseudo-likelihood ratio has the Wilks property as the true likelihood ratio does. Simulated examples are presented to evaluate the accuracy of the proposed semiparametric MLE method.
Subjects: 
information bound
kernel density estimator
maximum likelihood estimate
nonlinear regression
semiparametric model
U-statistic
Wilks property
JEL: 
C14
Document Type: 
Working Paper

Files in This Item:
File
Size
482.91 kB





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