power.u.test: sample size formula for the (one-sided) Mann-Whitney U test

In floatofmath/adaperm: Permutation tests for adaptive designs

Usage

power.u.test(res, propn = 1/2, sig.level = 0.025, power = 0.8,
  silent = F)

Arguments

res	relative effect size, i.e. probability that Y_i > X_i
propn	proportion of control group samples among total sample
sig.level	significance level
power	desired power
silent	should explanations be suppressed
n	total sample size

Value

sample size (rounded to the next larger integer)

Author(s)

Florian Klinglmueller

References

Noether, Gottfried E. "Sample size determination for some common nonparametric tests." Journal of the American Statistical Association 82.398 (1987): 645-647.

Examples

power.z.test(delta=1,sd=1,power=.84)
power.t.test(delta=1,sd=1,power=.84)
power.u.test(delta2res(delta=1,sd=1),power=.8)

floatofmath/adaperm documentation built on May 16, 2019, 1:18 p.m.