Versatile Sample-Size Calculation using Simulation

Abstract

I present a new Stata command, simsam, that uses simulation to determine the sample size required to achieve a given statistical power for any hypothesis test under any probability model that can be programmed in Stata. simsam returns the smallest sample size (or smallest multiple of 5, 10, or some other user-specified increment) so that the estimated power exceeds the target. The user controls the precision of the power estimate, and power is reported with a confidence interval. The sample size returned is reliable to the extent that if simsam is repeated, it will, nearly every time, give a sample size no more than one increment away.