6
$\begingroup$

Suppose I have a sample from a multinomial distribution $\text{Multi}(p_1,\ldots,p_k)$, where $p_1+\dots+p_k=1$. I'd like simultaneous confidence intervals for the $p_i$'s. That is, for any $\alpha\in(0,1)$, I want to define $\ell_i$'s and $u_i$'s (random variables depending on the random sample) such that $$ P(p_1\in [\ell_1,u_1],\ldots,p_n\in[\ell_n,u_n])\ge 1-\alpha, $$ and obviously the closer the left hand side is to $1-\alpha$, the better. The special consideration is that the number of classes won't be too small (e.g. 10-100 classes), so I'm wondering if there's something significantly better than a Bonferroni approach.

Thanks!

$\endgroup$
  • 1
    $\begingroup$ It's a reasonable question but the sense of 'confidence band' needs to be made more precise. What properties do you need it to have? E.g., should there be $1-\alpha$ confidence that every $p_i$ lies within $[l_i,u_i]$? Or maybe that at least one $p_i$ lies in its interval? Or perhaps should each $[l_i,u_i]$ be a $1-\alpha$ CI for $p_i$? Or maybe you would prefer the bounding box for a $1-\alpha$ simultaneous CI for all the $p_i$? $\endgroup$ – whuber Jan 4 '12 at 15:46
  • $\begingroup$ Thanks @whuber. I forgot the notion of simultaneous confidence intervals already exists. I've simplified my question statement significantly. $\endgroup$ – DavidR Jan 4 '12 at 20:21
6
$\begingroup$

Glaz and Sison (Journal of Statistical Planning and Inference, 1999) contains formulae for the Sison and Glaz confidence intervals for the MLE, which simulation showed perform quite well, and also some parametric bootstrap confidence intervals, also for the MLEs. I won't try to reproduce the math here, since there's rather a lot of it and it's in the paper anyway.

$\endgroup$
  • $\begingroup$ @whuber = thanks, I appreciate it. I'd missed the exclamation point when I tried to fix it up. $\endgroup$ – jbowman Jan 4 '12 at 23:32
  • $\begingroup$ The link is broken again $\endgroup$ – uLoop May 15 at 19:20
  • $\begingroup$ @uLoop - eventually all links break; that's why we give the original reference as well. I'll see if I can dig up an alternative, my first try didn't turn up anything. $\endgroup$ – jbowman May 15 at 19:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.