This paper deals with the calculation of large sample sizes for which confidence intervals of given widths for k multinomial frequencies are simultaneously correct at a given significance level. It is shown that a method for determining such sample sizes yields less conservative results than those given by other methods in use. Also, a less conservative empirical bound for sample sizes is proposed for confidence intervals of the same width.
Angers, C. 'A Graphical Method to Evaluate Sample Sizes for the Multinomial Distribution ." Technometrics16:3 (August 1974), 469-471.
2.
Bailey, B.J.R. "Large Sample Simultaneous Confidence Intervals for the Multinomial Probabilities Based on Transformations of the Cell Frequencies." Technometrics22:4 (November 1980), 583-589.
Hurtubise, R.A. "Sample Sizes and Confidence Intervals Associated with a Monte Carlo Simulation Model Possessing a Multinomial Output." Simulation12:2 (February 1969), 71-77.
5.
Mamrak, S.A. and Amer, P.D. "Estimating Confidence Intervals for Simulations of Computer Systems." Simulation35:6 (December 1980), 199-205.
6.
Miller, R.P.Simultaneous Statistical Inference. McGraw-Hill, New York, New York (1966).
7.
Tortora, R.D. 'A Note on Sample Size Estimation for Multinomial Populations." The American Statistician32:3 (August 1978), 100-102.
