Re: st: RE: fixed, random effects

[Raphael Schoenle <schoenle@Princeton.EDU>]

great, I am so happy that it runs -

one last thing - I try to store the estimates to run a hausman test  
(the goal of this exercise) but I get this:

. rfregk invest mvalue [aw = kstock]
                                                 Random-effects GLS  
regression
sd(u_company)                =  86.28686               Number of obs =   
   200
sd(e_company_t)              =  120.1563                           n =   
    10
sd(e_company_t + u_company)  =  147.9289                           T =   
    20

corr(u_company, X)           =  0 (assumed)            R-sq within   =   
0.6344
                                                            between  =   
0.9001
                                                            overall  =   
0.8283

                                                       chi2(  1)     =   
378.90
(theta = 0.7027)                                         Prob > chi2 =   
0.0000

------------------------------------------------------------------------ 
------
      invest |      Coef.   Std. Err.      z    P>|z|     [95% Conf.  
Interval]
------------- 
+----------------------------------------------------------------
      mvalue |   .2393068    .012294    19.47   0.000     .2152111     
.2634026
       _cons |  -123.7867    37.4564    -3.30   0.001    -197.1999    
-50.37351
------------------------------------------------------------------------ 
------

. est store RE
last estimation results not found, nothing to store
r(301);

is there a way to extract the var-cov as well as coefficient matrix for  
the hausman test?

many thanks,

-Raphael


On Nov 28, 2005, at 11:05 AM, Scott Merryman wrote:


> The problem is a few of lines in -rfregk- are wrapped.
>
> Lines 99 - 108 read:
>
> if "`weight'"!="" {
> 	by `ivar': replace ``i'' = /*
> 	*/
> 	cond(`dup',sum(`w'[_n-1]*``i''[_n-1])/s
> 	um(`w'[_n-1]),``i'')
> }
> else {
> 	by `ivar': replace ``i'' = /*
> 	*/
> 	cond(`dup',sum(``i''[_n-1])/(_n-1),``i'
> 	')
> }
>
> Either replace the -by `ivar' ...- lines so that the entire statement  
> is on
> one line or use /// to connect the lines:
>
> if "`weight'"!="" {
> 	by `ivar': replace ``i'' = ///
> 	cond(`dup',sum(`w'[_n-1]*``i''[_n-1]) ///
> 	/sum(`w'[_n-1]),``i'')
> }
> else {
> 	by `ivar': replace ``i'' = ///
> 	cond(`dup',sum(``i''[_n-1])/(_n-1),``i'')
> }
>
> Save the file, -discard-, and try again.
>
>
> Scott
>
>
> > -----Original Message-----
> > From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> > statalist@hsphsun2.harvard.edu] On Behalf Of Raphael Schoenle
> > Sent: Monday, November 28, 2005 9:33 AM
> > To: statalist@hsphsun2.harvard.edu
> > Subject: Re: st: RE: fixed, random effects
> >
> > Hi Scott,
> >
> > I just tried to run the example you sent me- does it work on your
> > computer? (i have stata se/8.2) I get an error message as below.
> > Thank you very much for the valuable hint!
> >
> > Best,
> >
> > Raphael
>
> *
> *   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/
*
*   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/