SIMULTANEOUS CONFIDENCE-INTERVALS AND SAMPLE-SIZE DETERMINATION FOR MULTINOMIAL PROPORTIONS

Simultaneous confidence interval procedures for multinomial proportion s are used in many areas of science. In this article two new simultane ous confidence interval procedures are introduced. Numerical results a re presented to evaluate these procedures and compare their performanc e with established methods that have been used in statistical literatu re. From the results presented in this article, it is evident that the new procedures are more accurate than the established ones, where the accuracy of the procedure is measured by the volume of the confidence region corresponding to the nominal coverage probability and the prob ability of coverage it achieves. In the sample size determination prob lem, the new procedures provide a sizable amount of savings as compare d to the procedures that have been used in many applications. Because both procedures performed equally well, the procedure that requires th e least amount of computing time is recommended.