Öhagen Patrik <Patrik.Ohagen <at> mpa.se> writes:

> 
> 
> Dear List,
> 
> I was asked to calculate a confidence interval for p*p. Is there any standard
techniques for calculating
> such an interval? Delta Method?
> 
> Thank you in advance!
> 


  if p is a generic value (i.e. not a probability) and you know the
variance (and are willing to assume normality) then you can indeed
use the delta method; there are a variety of other techniques
if you have the original data: fitting profile confidence limits, various
resampling methods including bootstrapping, etc.
(See section 6 of chapter 7 at http://www.zoo.ufl.edu/emdbook for
more details if you like).

  Ben Bolker

______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to