st: Thread-Index: AczTeoq+YhvLGaSRTF+DbDk6lqFX1w==

["STANEVA A. (497186)" <A.V.STANEVA.497186@swansea.ac.uk>]

Dear Statalist users,

I want to run the quantile regression in a way that the previously scalar intercept to be included. So, instead the intercept which automatically goes with the QR I need to replace it with a different one, let’s say consistently estimated or ‘correct’ one.

I am using Buchinsky (1998) correction of sample selection in the context of quantile regression, which requires higher order selection terms be included in the earnings function. There is however, the added complication that the constant term in the quantile wage regression is not identified in this instance, as it is conflated with the constant term of the higher order series in mills ratio used to control for sample selection. Therefore, I want to ‘plug’ in the ‘correct’ constant term into my qunitle regression.

Could someone help me with a bit programming how can I estimate this? I did try with a constraint option but it doesn’t seem to work into QR.

Many thanks in advance!

Anita

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/