Re: st: Subtracting out many fixed effects

[Austin Nichols <austinnichols@gmail.com>]

Chris Parker <cparker.phd2007@london.edu> :
 findit felsdvreg
etc.

On Mon, Jun 21, 2010 at 4:51 AM, Chris Parker
<cparker.phd2007@london.edu> wrote:
> Hi Clive,
>
> Thanks for the suggestion.  I'd always used i.MYVARIABLE in the
> regression explicitly which caused me problems with matsize.  After
> your response I dug in a bit deeper and found the -i()- option.
> Thanks for that.  However, the way I subtract out averages by male and
> then run the regression is exactly what the -i()- option should do.
> This re-creates the regression as I would expect.  The problem I run
> into is then subtracting averages from a second categorical variables
> (in my example, topper).  I can't specify two or more variables with
> the -i()- option so I'm back to the same issue as before.
>
> In addition, the group fixed effects are not the fixed effects I am
> trying to manually subtract out.  I let Stata handle those.
>
> Chris
>
> On Mon, Jun 21, 2010 at 9:10 AM, Clive Nicholas
> <clivelists@googlemail.com> wrote:
>> Chris Parker wrote:
>>
>>> I'm trying to run a panel regression with a large number of fixed
>>> effects in addition to the group fixed effects that xtreg takes care
>>> of for me.  Because of the large number of variables I cross the limit
>>> on matsize of 11,000.  I have therefore opted to subtract out some of
>>> the fixed effects.  My problem is that I don't know the best way to do
>>> this when there are multiple fixed effects to be subtracted out.
>>
>> [...]
>>
>> You ought to be running -xtreg- with the -i()- and -fe- options
>> switched on: this will automatically take care of your fixed effects
>> and do the de-meaning for you. What happens when you do this?
>>
>> --
>> Clive Nicholas

*
*   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/