We propose a root n consistent estimator for B0 when the qth conditional quantile of Y given X=x and Z=z takes the semi linear form g(x)+z'B0 where g(.) is an un- known real valued function,B0 a finite dimensional parameter and (X,Z) a couple of explanatory variables. Importantly, our estimator attains,under homoscedasticity,the semi parametric efficiency bound. This estimation is conducted in two steps. First, a Robinson's like demeaning of the original model is employed which provides a new quantile regression whose nuisance terms are estimated via a non parametric proce- dure.In the second stage, a quantile regression is conducted by smoothing the check function. We show that the previous estimator belongs to a class of estimators we propose to name "two stage semi parametric quantile".
Download Info
To download:
If you experience problems downloading a file, check if you have the
proper application to
view it first. Information about this may be contained
in the File-Format links below. In case of further problems read
the IDEAS help
page. Note that these files are not on the IDEAS
site. Please be patient as the files may be large.
Publisher Info
Paper provided by Department of Economics, Louisiana State University in its series Departmental Working Papers with number
2009-05.