Re: Bootstrap confidence intervals

Keith O'Rourke (orourke@utstat.toronto.edu)
Wed, 3 Aug 94 10:00:26 EDT


>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.

>(B) (to get a CI) The underlying population distribution is not
> necessarily the same as the observed sample distribution.

But, there are many more kinds of bootstrap

Non-parametric - treat the sample as if it were the population

Semi-parametric - treat some aspects of the sample as if ...

Parametric a: Discretize a parametric distribution so it can be boot strapped
b: Draw random samples from a parametric distribution (simmulate)

For quite thorough but not too technical account of these issues

Efron & Tibshirani, An Introduction to the Bootstrap, Chapman & Hall 1993