Re: st: Construct Null Datasets through Bootstrap Resampling

["Michael Blasnik" <michael.blasnik@verizon.net>]

One more difference between the approaches that you should recognize is that 
bsample samples with replacement (a value from a given observation can 
appear more than once) while the scramble program I wrote does not -- it 
simply re-arranges the order.  I'm not sure which is preferred for your 
application.

If you end up wanting to sample without replacement (as scramble does) but 
want to keep groups of variables together, a relatively modest change to the 
scramble code would do the trick.

Michael Blasnik

----- Original Message ----- 
From: "Erik Ingelsson" <Erik.Ingelsson@pubcare.uu.se>
To: <statalist@hsphsun2.harvard.edu>
Sent: Thursday, November 30, 2006 11:37 PM
Subject: Re: st: Construct Null Datasets through Bootstrap Resampling



> Dear Austin,
>
> Well, actually Michael Blasniks suggestion was a bit closer to what I 
> first intended. But when looking at your suggestion and reading some 
> literate on multiple testing in genetic analyses, I realized that I 
> actually perhaps should not scramble all variables, but rather keep  them 
> together in two groups (e.g. keeping var 1-4 togeter as in the  orignal 
> observation, but randomly pairing them with var 5-8). For  this, your 
> solution worked perfectly. The next step is to meet with  our senior 
> statistician and show him what I got. He can tell which way  to go, I 
> hope.
>
> Thanks a lot to both of you! I am impressed by the fast answers!
> Erik Ingelsson
*
*   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/