Statalist


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

st: re: Wald test across equations after ivregress or ivreg2


From   Kit Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: re: Wald test across equations after ivregress or ivreg2
Date   Fri, 24 Jul 2009 09:13:34 -0400

<>
Godiva wanted to do
ivregress y (a = x y z) b c
est store one
ivregress y (a = x y z) b c d
est store two
 suest one two

Actually this is not workable code, as -ivregress- (unlike -ivreg2- from SSC or -ivreg-) requires a second argument (such as 2sls). But you don't need to do the test to determine whether the coefficients of 'a' are the same. If the coefficient of 'd' is significantly different than zero in the second regression, then the first regression is misspecified, and should be discarded: and vice versa. As the second model nests the other, there is no need to consider them as competing models, as one of them must be misspecified.

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/



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