Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Re: rolling regression


From   Kit Baum <baum@bc.edu>
To   "Dr.K.Hassan Naqvi" <khnaqvi@gmail.com>
Subject   st: Re: rolling regression
Date   Thu, 30 Oct 2008 06:09:53 -0400

I don't see why you need -nlcom- to calculate standard errors, unless you are trying to compute nonlinear combinations of the coefficients. - rolling- will calculate standard errors. Modifying the second example in its helpfile,


        . webuse lutkepohl2, clear
        . tsset qtr
. rolling _b _se, window(10): regress dln_inv dln_inc dln_consump

If for some reason you would like to compute nlcoms of the estimated coefficients with their associated standard errors, you would need to write a little 'wrapper' program that generated the desired results and returned them to -rolling-, and execute that rather than - regress-. That would be a bit more work. Please clarify what you are trying to do.

Kit

Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


On Oct 30, 2008, at 24:52 , Dr.K.Hassan Naqvi wrote:

Dear Sir,

I want to have the rolling coefficent alongwith standad errors . I have been able to estimate rolling regression using stata 9 form its dropdown manue. However when I tried to estimate the standard errors of the rolling coefficents using nlcom command, it does not work and always given a message "last estimation not found".

Based on your knowledge, may I request you to please guide me how can I apply nlcom on rolling regression in stata 9.

thanks,
Hassan

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index