Re: st: problem with generated regressands and WLS

[Austin Nichols <austinnichols@gmail.com>]

Arka Roy Chaudhuri <gabuisi@gmail.com>:
If you think beta is measured with an independent error, i.e. no
endogeneity or other endemic problems, you can ignore the fact that it
is generated; measurement error in the depvar is usually not a
problem. But perhaps you are looking for improved efficiency, and you
want to use the squared SE on beta as a measure of the error
variance--but it does not vary by observation--see the manual entry on
-vwls- for example.  Is your "second stage" in matrix form using the
same y and x and so forth, or have you reused notation?

2010/10/12 Arka Roy Chaudhuri <gabuisi@gmail.com>:
> Hi,
>
>   I have a two-stage estimation strategy. In the first stage I
> estimate a no of parameters β1.....βN. I use these parameters as my
> dependent variable/regessand in the second stage equation.Now since my
> regressands are generated I need to use the first stage
> variance-covariance matrix of my variables as weights in the second
> stage.To elaborate here is my first stage:
>
>  y= α + β1x1+ β2x2 +................. +βnxn+ ρ1z1 +  ρ2z2 + u
>
> Now I want to use the β's from the first stage as my second stage
> regressands. My second stage looks like:
>
>  β= γ+ δx + ε
>
> Here in the second stage as my β's are estimates from the first stage
> I believe I should use the variance-covariance matrix of the β's as
> the weighing matrix in the second stage.
> Could somebody please tell me if I am doing this correctly and how to
> implement it in Stata?Thanks
>
> Arka
>
> --
> Arka Roy Chaudhuri
> PhD Student
> University of British Columbia
>

*
*   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/