Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Wanted: simulated data with cluster structure


From   Austin Nichols <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Wanted: simulated data with cluster structure
Date   Wed, 11 May 2011 12:06:03 -0400

Bert Jung <bjung59@gmail.com>:
Don't forget that your X vars (if any) also must be clustered if you
want the nonrobust SE to be incorrect on average.

On Wed, May 11, 2011 at 8:04 AM, Bert Jung <bjung59@gmail.com> wrote:
> Thanks Stas... -expand-, of course...
> Bert
>
> On Tue, May 10, 2011 at 10:48 PM, Stas Kolenikov <skolenik@gmail.com> wrote:
>> On Tue, May 10, 2011 at 9:31 PM, Bert Jung <bjung59@gmail.com> wrote:
>>> I need a simulated dataset with clustered data to test a user-written
>>> bootstrap routine.  The only important feature is that the cluster
>>> structure should lead to visibly biased standard errors in -regress-
>>> and that a -regress, cluster()- should return the correct standard
>>> error.
>>
>> Oh come on.
>>
>> set obs 20
>> gen u = rnormal()
>> gen cl = _n
>> expand 20
>> gen e = 0.2*rnormal()
>> gen y = u+e
>> reg y
>> reg y, cl(cl)
>>
>> --

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index