A SAS® macro for constructing simultaneous confidence intervals for multinomial proportions
Quesenberry and Hurst (1964), Goodman (1965) and Fitzpatrick and Scott (1987) proposed simultaneous construction of confidence intervals for multinomial proportions, however, statistical computing packages do not generally give one the option of specifying the type of construction to be used. We have written a SAS macro using PROC IML that takes multinomial cell counts as input and returns simultaneous confidence intervals with the user-specified coverage probability. Two main features of the macro are its ease of use and its flexibility in allowing the user to choose among six methods of constructing confidence intervals for multinomial proportions. Based on simulation May and Johnson (1997) recommended the intervals proposed by Goodman (1965) in most practical applications.
Keywords: Categorical data; Multinomial distribution; Simultaneous interval estimation
You can download the .pdf of this paper from ResearchGate.
This paper is cited on the SAS blog post Simultaneous confidence intervals for multinomial proportions.