Re: st: bootstrap with overlapping clusters

[Steven Samuels <ssamuels@albany.edu>]

Nick, please provide more details about your problem:  I'd like to  
know: what the target population is, how the sample was drawn, what  
defines the clusters, and in what way they overlap. Also, what  
questions do you want to answer with the bootstrap--do you want  
simple descriptive statistics for subpopulations or do you want to do  
make comparisons, do modeling, or all of the above.  If you are  
working with an established survey, is there a document which  
suggests an analysis setup (not necessarily in Stata)?

I suspect that you have a panel study in which different households?  
firms? rotate in and out, so that the overlap is in time--but I await  
your answer.

Steve

  On May 23, 2007, at 11:59 AM, Nicholas.Kraninger@bos.frb.org wrote:


> Is there a way to use the cluster option on the bootstrap command  
> to take
> draws of overlapping clusters?
>
> Thanks,
> Nick
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/