Surely the bootstrap intervals will just give you the sample order
statistics x(i) and x(N-i+1) for a sample of size N. You can do a
simple(ish) calculation to find P[median < x(i)] and save some
computer cycles
On Tuesday, Mar 18, 2003, at 14:45 Europe/London, PEP PUIGVERT wrote:
> I agree with the last answer: bootstrap methods are a possible tool to
> find
> a good CI for the median. And Introduction to the Bootstrap by Efron
> and
> Tibshirani is a good introduction in this topic.
> Kind regards. Pep.
>
>
>> From: jorge villacampa Reply-To: jorge villacampa To:
>> [log in to unmask] Subject: Confidence intervals for medians
>> Date: Tue,
>> 18 Mar 2003 08:49:29 +0000
>>
>> Dear all,
>>
>> I would like to do a confidence interval for the median. could you
>> tell me
>> any website, book where to find it?
>>
>> Thanks a lot
>>
>> Kind regards
>>
>> Jorge
>>
>>
>>
>>
>> --------------------------------- With Yahoo! Mail you can get a
>> bigger
>> mailbox -- choose a size that fits your needs
>
> _________________________________________________________________
> The new MSN 8: smart spam protection and 2 months FREE*
> http://join.msn.com/?page=features/junkmail
>
|