Estimating confidence limits for p from a sample of a binomial distribution

All topics, except for ones related to water fights. Posts made here do not count towards your post count.
Locked
User avatar
cantab
Posts: 1492
Joined: Fri Oct 19, 2007 1:35 pm

Estimating confidence limits for p from a sample of a binomial distribution

Post by cantab » Fri Jul 25, 2008 10:12 am

OK any statisticians here: stats question.

I have a sample from a binomial distribution. p (probability of success) of the distribution is unknown. Obviously I can estimate p from the sample p, but I need confidence limits on that estimate. I'm not sure how to do that.

EDIT: To make things awkward, the number of successes is zero. The sample size is 45, 26 or 8

Followup question. If I have two samples, again from binomial distributions with unknown p, what is the probability that they come from the same population. One sample has sample p = 0

Context: The first problem relates to estimating the probability of making sales based on a first day of zero sales. The second relates to figuring out if a day of training actually makes a difference or if it's just luck. The results will determine whether or not I continue with the job.
Last edited by cantab on Fri Jul 25, 2008 10:50 am, edited 1 time in total.

Locked