A subroutine is presented to calculate exact confidence limits for the number of successes in a finite population, given a random sample drawn without replacement. The limits are calculated recursively from a series of hypergeometric probability distributions. The use of arbitrary starting values in each recursion eliminates the computation of initial probability values based on cumbersome factorial expressions.
Get full access to this article
View all access options for this article.
References
1.
Buonaccorsi, J. P. (1987). A note on confidence intervals for proportions in finite populations. The American Statistician, 41, 215-218.
