Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

st: How to implement FGLS on estimated regression coefficients?


From   Michael Boehm <michael.boehm1@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: How to implement FGLS on estimated regression coefficients?
Date   Fri, 17 Feb 2012 00:00:39 +0000

Dear all,

I want to run a FGLS estimation on the coefficients from a seemingly
unrelated regression, ie I first regress

"sureg (dd i.decade##(c.t1 c.t2) ) (w i.decade##(c.t1 c.t2)), coeflegend"

with output

"			
                 |      Coef.          Legend
-------------+----------------------------------------------------------------
    dd           |
    1.decade |   .1909903       _b[dd:1.decade]
               t1 |   .1976521      _b[dd:t1]
               t2 |  -.2013332      _b[dd:t2]
                   |
 decade#c.t1 |
                1  |  -.0220626     _b[dd:1.decade#c.t1]
                   |
 decade#c.t2 |
               1  |   .0245381      _b[dd:1.decade#c.t2]
                   |
         _cons |   .5002711       _b[dd:_cons]
-------------+----------------------------------------------------------------
w                |
    1.decade |   .5814188        _b[w:1.decade]
          t1      |    1.50046        _b[w:t1]
          t2      |   1.497409       _b[w:t2]
                   |
 decade#c.t1 |
          1        |   .1770365       _b[w:1.decade#c.t1]
                    |
 decade#c.t2 |
          1        |  -.1983975       _b[w:1.decade#c.t2]
                    |
       _cons    |   1.809371      _b[w:_cons]
"

Then I want to stack y=(_b[w:1.decade#c.t1]  _b[w:1.decade#c.t2])' and
FGLS-regress it on x=(_b[dd:t1] _b[dd:t2])' (a regression with two
observations) with the weighting matrix being the covariance matrix of
these parameter estimates from the previous SUR regression. Ideally, I
would also like to report the value of the objective function that
FGLS minimizes in optimum, because this is supposed to be chi-squared
distributed with 2 degrees of freedom under the H0 that my model is
true.

Sorry for this long explanation, but does anyone know how to implement
this procedure nicely in Stata?

Thanks,
Michael
*
*   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   |   Site index