Re: st: FGLS Estimation

[Tirthankar Chakravarty <tirthankar.chakravarty@gmail.com>]

You may want to try -vwls-

T

2010/6/9 Stefan <stefan@gravemeyer.de>:
> Hello,
>
> I am trying to do a three step FGLS estimation with Stata but run into some
> problems maybe someone has a good idea.
> I start with an ordinary OLS regression and then regress the residuals on
> the same right hand side variables.
> Then I have adjust the right and left hand side variables as well as the
> residuals by the predicted values of the right hand side variable y.
> Here is my problem:
> I did not find a way to adjust the residuals for the regression by my
> predicted weights.
>
> This is the transformation I would like to do:
> y/weight=X/weight * ß + e/weight
>
> Adjusting the variables is no problem.  I simply use egen to generate new
> ones.
> (using aweights does not work since I can not simply square my predicted
> value weight to fit the sqrt(weight)*y formula used by aweights, maybe there
> is exists an easier way here too?)
> But I simply do not know how to transform the residuals for my next step
> regression.
>
> Thanks in advance
>
> Stefan
>
>
> *
> *   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/
>



-- 
To every ω-consistent recursive class κ of formulae there correspond
recursive class signs r, such that neither v Gen r nor Neg(v Gen r)
belongs to Flg(κ) (where v is the free variable of r).

*
*   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/