[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
"FEIVESON, ALAN H. (AL) (JSC-SK) (NASA)" <alan.h.feiveson@nasa.gov> |

To |
"'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |

Subject |
st: RE: Binomial confidence intervals (more) |

Date |
Thu, 9 Sep 2004 08:14:54 -0500 |

All this discussion about failure of binomial confidence intervals to give "exact" coverage also applies to the Fisher "exact" test, whose actual level (probability of rejecting the null hypothesis of equal proportions, when in fact the proportions are equal) is usually less than the nominal level, depending on the true proprtions. In the frequentist setting, it's the same problem - there are only a finite number of possible outcomes. Al Feiveson -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of Joseph Coveney Sent: Wednesday, September 08, 2004 11:56 PM To: Statalist Subject: Binomial confidence intervals (more) Nick Cox wrote: A small tweak to Joseph's useful program. It feeds on a given sample size and population proportion, but irrespective of the latter 0.5 is wired in to a crucial line. So I suggest return scalar `method'_covered = (`pi' >= r(lb)) & (`pi' <= r(ub)) -inrange(`pi', r(lb), r(ub))- is a another way of doing it. ---------------------------------------------------------------------------- Whoops--thanks Nick! It started out fixed at 0.5 throughout; it originally was intended only as an illustration where listers could see in the Stata code (invoking the definition of the term, confidence interval) the difference between probability of coverage, which was Bobby Gutierrez's point, and *exact* in Constantine Daskalakis's sense. For the purpose of the illustration, the true proportion was arbitrarily fixed at 0.5 in the original code. When Paul Seed posted his question, the original purpose was abandoned, and I tried--alas, too hastily--to generalize the program for another purpose: a further illustration of the versatility of simulation to answer questions in the absence of alternative means. Thanks also (and to Bobby) for making -bincoverage- available as one such alternative means. It is certainly preferable to simulation for practical work. Joseph Coveney * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**Re: st: RE: Binomial confidence intervals (more)***From:*Richard Williams <Richard.A.Williams.5@nd.edu>

- Prev by Date:
**st: RE: outtex and append** - Next by Date:
**Re: st: RE: Binomial confidence intervals (more)** - Previous by thread:
**st: RE: outtex and append** - Next by thread:
**Re: st: RE: Binomial confidence intervals (more)** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |