On Mar 7, 2011, at 19:12 , Bentley Coffey wrote:
> Just to tie up this thread, I wanted to report my solution:
>
> When (n-1)p is an integer, there is a closed form solution:
> pbinom(j-1,n,...)
>
> When it is not an integer, its fairly easy to approximate the solution by
> interpolating betwee
Just to tie up this thread, I wanted to report my solution:
When (n-1)p is an integer, there is a closed form solution:
pbinom(j-1,n,...)
When it is not an integer, its fairly easy to approximate the solution by
interpolating between the closed-form solutions: fitting log(1 - probability
from clo
Duncan,
I'm not sure how I missed your message. Sorry. What you describe is what I
do when (n-1)p is an integer so that R computes the sample quantile using a
single order statistic. My later posting has that exact binomial expression
in there as a special case. When (n-1)p is not an integer then
rneysville WV, 25430
> (304) 724-4480
> "Is the room still a room when its empty? Does the room,
> the thing itself have purpose? Or do we, what's the word... imbue it."
> - Jubal Early, Firefly
>
> r-help-boun...@r-project.org wrote on 02/14/2011 09:58:09 AM:
On 14/02/2011 9:58 AM, Bentley Coffey wrote:
I need to calculate the probability that a sample quantile will exceed a
threshold given the size of the iid sample and the parameters describing the
distribution of each observation (normal, in my case). I can compute the
probability with brute force
ly
r-help-boun...@r-project.org wrote on 02/14/2011 09:58:09 AM:
> [image removed]
>
> [R] CDF of Sample Quantile
>
> Bentley Coffey
>
> to:
>
> r-help
>
> 02/14/2011 01:58 PM
>
> Sent by:
>
> r-help-boun...@r-project.org
>
> I need
I need to calculate the probability that a sample quantile will exceed a
threshold given the size of the iid sample and the parameters describing the
distribution of each observation (normal, in my case). I can compute the
probability with brute force simulation: simulate a size N sample, apply R's
7 matches
Mail list logo
