power.fisher.test {statmod}

Power of Fisher's Exact Test for Comparing Proportions

Description

Calculate by simulation the power of Fisher's exact test for comparing two proportions given two margin counts.

Usage

power.fisher.test(p1, p2, n1, n2, alpha=0.05, nsim=100, alternative="two.sided")

Arguments

p1

first proportion to be compared.

p2

second proportion to be compared.

n1

first sample size.

n2

second sample size.

alpha

significance level.

nsim

number of data sets to simulate.

alternative

indicates the alternative hypothesis and must be one of "two.sided", "greater" or "less".

Details

Estimates the power of Fisher's exact test for testing the null hypothesis that p1 equals p2 against the alternative that they are not equal.

The power is estimated by simulation. The function generates nsim pairs of binomial deviates and calls fisher.test to obtain nsim p-values. The required power is tnen the proportion of the simulated p-values that are less than alpha.

Value

Estimated power of the test.

Author(s)

Gordon Smyth

See Also

fisher.test, power.t.test

Examples

power.fisher.test(0.5,0.9,20,20) # 70% chance of detecting difference