> The idea of using the bootstrap to obtain CI's has always
> struck me as perverse, since one is simultaneously assuming:
>
> (A) (for bootstrapping) The underlying population distribution is
> precisely the same as the observed sample distribution.
I don't think this is what we are assuming. I like the
explanation in Peter Hall's recent book on the bootstrap
(I don't recall the exact title). Bootstrapping is based
on an analogy:
Bootstrap sample is to original sample
as
original sample is to population.
Peter Hall gives an example using Russian dolls
to illustrate this.
>
> (B) (to get a CI) The underlying population distribution is not
> necessarily the same as the observed sample distribution.
>
But this is accounted for in the calculation of the CI.
*******************************************************************
Michael P. Cohen Internet: xic@cu.nih.gov
National Center for Education Statistics
555 New Jersey Avenue NW #408
Washington, DC 20208-5654 USA
*******************************************************************