Statalist


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

Re: st: bootstrap with cluster


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: bootstrap with cluster
Date   Wed, 13 May 2009 09:23:37 -0500

No, the clusters are resampled as a whole. That's how it should be
done, at the very minimum, although the issue is far more complicated:
the variance estimates are still biased unless you do some extra work.
See the following references:
http://www.citeulike.org/user/ctacmo/article/1334050,
http://www.citeulike.org/user/ctacmo/article/582039,
http://www.citeulike.org/user/ctacmo/article/4510149 (although the
authors of the latter paper are totally unaware that this issue has
been studied in survey statistics literature for at least 20 years).

On 5/13/09, Gianluca Cafiso <gcafiso@unict.it> wrote:
> Dear Statalisters,
>
>  I am facing again the lack of Stata manuals...
>
>  When the "cluster" option in the "bootstrap" command is specified, how is
> the resampling implemented?
>
>  My guess is that resampling occurs only within clusters. Namely, that given
> N observed data which constitute the sampe and r clusters in which the N
> observations are grouped, r simultaneous resamplings occur (each B times,
> number of replications specified) where the resampling for the rth cluster
> uses only observations which are included in the cluster. Am I right?
>
>  Many thanks for any contribution on this.
>
>  (any reference on this is very apprecciated as well)
>
> ___________________________________________________________
>  Gianluca Cafiso
>  Research Fellow, Economics Department-University of Catania.
>  Corso Italia 55, Catania, Italy.
>
>  e-mail: gcafiso@unict.it
>  tel.:  +39 0957537745
>
>
> ----------------------------------------------------------------
>  Università di Catania - C.E.A.
>  Servizio di Posta Elettronica
>  http://www.cea.unict.it
>
>
>  *
>  *   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/
>


-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   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   |   What's new   |   Site index