Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: URGENT! testing the equality of coefficients across quantiles


From   "Michael Blasnik" <michael.blasnik@verizon.net>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: URGENT! testing the equality of coefficients across quantiles
Date   Fri, 15 Oct 2004 10:10:53 -0400

The answer is that the test provided is what you want -- a test of equality
of all coefficients across the models.  You can check this by using multiple
test statements with the accum option and see that you get the same results
regardless of which equations are used for the reference for each of the 4
tests as long as each equation is there somewhere

By the way, please don't use URGENT in your subject, it led me delay my
reply...

Michael Blasnik
michael.blasnik@verizon.net


----- Original Message ----- 
From: "fatma bircan" <bircan@metu.edu.tr>
To: <statalist@hsphsun2.harvard.edu>
Sent: Friday, October 15, 2004 5:00 AM
Subject: Re: st: URGENT! testing the equality of coefficients across
quantiles


> -------------------
>
> Fatma Bircan
> Middle East Technical University
> Department of Economics
> 210-2032
>
> I am trying to perform a wald test to see if the parameters are equal
> across the specified quantiles after running a quantile regression.
>
> test command for after-a multi-equation-model performs this task. But
> when it is more than one equation, the equality of parameters across
> the equations is tested in relation to the first equation we put in
> the command line,
>
> Example,
> .
>  test y1 coefs = y2 coefs = y3 coefs
>         . test [y1=y2=y3]
>         . test [y1=y2=y3]: site1 site2 site3
>
> The result compares the coefs. in y1 and y2 and in y1 and y3, to test
> the equality for y2 and y3 we need to write another line, but then the
> number of restrictions and df changes, with 5 equations I need to
> write more than two lines. So I am confused about how to test the
> equlity of parameters in all five equations at the same time, or is it
> necessary to do so???
>
> Thank you
>
>


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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