[R] Calculating confidence in an estimate including numberof trials?

[Daniel Lakeland]

On Tue, Oct 16, 2007 at 04:44:54PM -0400, Zembower, Kevin wrote:
> Daniel, thanks for your suggestion. So, it's just done like this:
> > pnorm(21, mean=20, sd=10/4) - pnorm(19, mean=20, sd=10/4)
> [1] 0.3108435
> > # OR
> > pnorm(21, mean=20, sd=10/sqrt(16)) - pnorm(19, mean=20,
> sd=10/sqrt(16))
> [1] 0.3108435

You have calculated the probability that your sample estimate of the
mean will be between 19 and 21, in your previous post you asked for 10
to 21, but I am guessing it was simply a typo. In any case, it looks
like you are on the right track.


-- 
Daniel Lakeland
dlakelan at street-artists.org
http://www.street-artists.org/~dlakelan