st: RE: Robust std errors for fixed effects LOGIT

["Seed, Paul" <paul.seed@kcl.ac.uk>]

Dear Gou Chen, 

I'm copying this to Statalist, as Glauco's problem 
has already been discussed there.

The only vce options offered with xtlogit are 
oim, bootstrap, or jackknife, so my first solution does not work.
My mistake, I fear.

However, as you are using a fixed effect model, you can use 
logit and add your cluster variable as a fixed effect.
-logit-  does take cluster().  

Using your example;
logit union age grade not_smsa south i.black, cluster( black) nolog
You will see that you get the same estimates (but different SE)
as with 
xtlogit union age grade not_smsa south, robust

Here, because black is binary, there is just one extra term.
In more realistic examples, there will be a large number.


BW


From: Guo Chen [mailto:chenx711@umn.edu] 
Sent: 29 November 2011 01:47
To: Seed, Paul
Subject: Robust std errors for fixed effects LOGIT

Hi Paul:

I am using your method given in the post below to do the robust se for a fixed effects logit model. It seems xtlogit with robust option doesn't work for me... I spent the whole afternoon trying to figure out but failed.. Since you actively answered another related post I just want to direct the question to you..


. webuse union, clear
(NLS Women 14-24 in 1968)

. xtset black
       panel variable:  black (unbalanced)

. xtlogit union age grade not_smsa south, robust
option robust not allowed
r(198);


Do you know any other methods to do it?

Any input will be greatly appreciated. Sorry if it's not convenient..


Thanks!

Guo

~~~~~~~~~~~begin forwarding~~~~~~~~~~~~~~

http://www.stata.com/statalist/archive/2008-10/msg00656.html

Re: st: xtlogit with cluster
________________________________________
From
  Paul Seed <paul.seed@kcl.ac.uk>
To
  "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject
  Re: st: xtlogit with cluster
Date
  Mon, 13 Oct 2008 10:53:43 +0100
________________________________________
Dear Statalist, The solution to Glauco's problem might be to use the -robust- option. Like other -xt- commands, -xtlogit-


automatically custers by the panel identifier
(the -i- variable) when -robust- is called.

If he wants to cluster by something else, it would appear he has multi-level data; and (unless -xtmixed- now has a logistic- option) Stata might not be the best package.


Best Wishes,
Paul Seed

------------------------------

Date: Sun, 12 Oct 2008 00:01:40 +0200
From: "Martin Weiss" <martin.weiss1@gmx.de>
Subject: Re: st: xtlogit with cluster

What is the answer to Glauco`s question then? The -help- file you mention
says: " Not all the options documented below work with all xt estimation
commands". -h xtlogit- certainly does not mention a -cluster- option for the
se. It does link to -help vce_option- but only allows some of the choices
mentioned there. -cluster- does not seem to be one of them...

HTH
Martin
_______________________
- ----- Original Message -----
From: "Steven Samuels" <sjhsamuels@earthlink.net>
To: <statalist@hsphsun2.harvard.edu>
Sent: Saturday, October 11, 2008 11:53 PM
Subject: Re: st: xtlogit with cluster



> -help xt_vce_options-  has the answer.  I found this from -help xtlogit-
>
> -Steve
>
>
> On Oct 11, 2008, at 3:45 PM, Glauco Peres wrote:
>
>> Dear STATAlista,
>>
>> How do I run a xtlogit and add cluster command [vce(cluster)] in  order
>> to correct the standard errors?
>>
>> Thanks a lot,
>>
>> Glauco
>>
>>
>>
>>       Novos endereços, o Yahoo! que você conhece. Crie um email  novo com
>> a sua cara @ymail.com ou @rocketmail.com.
>> http://br.new.mail.yahoo.com/addresses
>>

*
*   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/