Re: st: Fixed effect model with weight: xtivreg2 [aweight],fe?

 From "Dohan Kim" To statalist@hsphsun2.harvard.edu Subject Re: st: Fixed effect model with weight: xtivreg2 [aweight],fe? Date Sat, 2 Aug 2008 20:44:59 -0400

```Scott,

Thank you very much.

Do Han.

On Fri, Aug 1, 2008 at 3:41 PM, Scott Merryman <scott.merryman@gmail.com> wrote:
> You could use -areg-  or -xtivreg2-
>
> For example:
>
> . webuse grunfeld
>
> . areg invest mval [aw = kstock], ab(com)
>
> Linear regression, absorbing indicators                Number of obs =     200
>                                                       F(  1,   189) =  327.94
>                                                       Prob > F      =  0.0000
>                                                       R-squared     =  0.9082
>                                                       Root MSE      =  120.16
>
> ------------------------------------------------------------------------------
>      invest |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
> -------------+----------------------------------------------------------------
>      mvalue |   .2971024   .0164063    18.11   0.000     .2647394    .3294654
>       _cons |  -226.6022   30.39757    -7.45   0.000    -286.5643   -166.6401
> -------------+----------------------------------------------------------------
>     company |         F(9, 189) =     18.251   0.000          (10 categories)
>
> . xtivreg2 invest mvalue [aw = kstock],  small fe
>
> FIXED EFFECTS ESTIMATION
> ------------------------
> Number of groups =        10                    Obs per group: min =        20
>                                                               avg =      20.0
>                                                               max =        20
>
> OLS estimation
> --------------
>
> Estimates efficient for homoskedasticity only
> Statistics consistent for homoskedasticity only
>
>                                                      Number of obs =      200
>                                                      F(  1,   189) =   327.94
>                                                      Prob > F      =   0.0000
> Total (centered) SS     =  7463301.849                Centered R2   =   0.6344
> Total (uncentered) SS   =  7463301.849                Uncentered R2 =   0.6344
> Residual SS             =   2728695.85                Root MSE      =    120.2
>
> ------------------------------------------------------------------------------
>      invest |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
> -------------+----------------------------------------------------------------
>      mvalue |   .2971024   .0164063    18.11   0.000     .2647394    .3294654
> ------------------------------------------------------------------------------
> Included instruments: mvalue
> ------------------------------------------------------------------------------
>
>
> Scott
>
>
> On Thu, Jul 31, 2008 at 11:50 PM, Dohan Kim <dongmool@gmail.com> wrote:
>> Hello Statalist members,
>> I'd like to run a fixed effect model with an analytic weight. "xtreg,
>> fe" doesn't allow a weight option, and some people in this forum
>> suggested to use xtivreg2 [aweight=],fe. Can I use "xtivreg2,fe" even
>> though I don't have any endogenous variables? In other words, can
>> "xtivreg2 [aweight=],fe" be an alternative to a simple fixed effect
>> model with a weight? If I can't use xtivreg2, are there any other ways
>> I can run a fixed effect model with an analytic weight? I'll
>>
>> --
>> Do Han Kim
> *
> *   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/
>

--
Do Han Kim
PhD Candidate,
Dept. of Public Administration and Policy
Rockefeller College, University at Albany
*
*   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/
```