Statalist


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

st: AW: Re: testing across different sub-samples


From   "Martin Weiss" <[email protected]>
To   <[email protected]>
Subject   st: AW: Re: testing across different sub-samples
Date   Fri, 25 Sep 2009 13:01:40 +0200

<> 



What would the test with one df look like in my example?



HTH
Martin

-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Kit Baum
Gesendet: Freitag, 25. September 2009 12:59
An: [email protected]
Betreff: st: Re: testing across different sub-samples

<>
Martin is quite correct in describing the test. However it is not the  
test the original poster asked for, which is a test with one d.f.  
(that the SUM of three coefficients in one model equals the SUM of  
three coefficients in another). Martin describes a much stronger  
statement: that EACH coefficient in one model equals its counterpart.  
That is a valid test, but a test with 3 d.f., not 1.

Kit

On Sep 25, 2009, at 2:33 AM, Martin wrote:

>
> This is a joint test for the equality of three coefficients across
> equations:
>
>
> **************
> sysuse auto, clear
> reg pr we le he if f
> est store for
> reg pr we le he if !f
> est store dom
> suest for dom
> test [for_mean=dom_mean]:weight length headroom
> **************


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

*
*   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/


*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index