PASSED (version 1.0-3)

power_Beta: Power Calculations for Test of Two Beta Means

Description

Compute the power for a test of two sample means with beta distributions, or determine the minimum sample size to obtain a target power.

Usage

power_Beta(n1 = NULL, n2 = NULL, power = NULL, sig.level = 0.05, 
mu1 = NULL, sd1 = NULL, mu2 = NULL, equal.sample = TRUE,
trials = 100, equal.precision = TRUE, sd2 = NULL, 
link.type = c("logit", "probit", "cloglog", "cauchit", "log", "loglog"))

Arguments

n1

sample size in group 1, or sample size in each group if equal.sample = TRUE

n2

sample size in group 2

power

power of test (1 minus Type II error probability)

sig.level

significance level (Type I error probability)

mu1

sample mean of group 1

sd1

standard deviation for group 1

mu2

sample mean of group 2

equal.sample

equal sample sizes for two groups, see details

trials

number of trials in simulation

equal.precision

equal dispersion parameter assumption in simulation

sd2

standard deviation for group 2. Only applicable when equal.precision = FALSE

link.type

type of link used in the beta regression, see details

Value

Object of class "power.htest", a list of the arguments (including the computed one) augmented with method and note elements.

Details

Exactly one of the parameters n1, n2 and power must be passed as NULL, and that parameter is determined from the others. This function allows you to set the number of trials in the simulation to control the result accuracy, and type of link used in the beta regression. You can choose one of the following: "logit", "probit", "cloglog", "cauchit", "log", "loglog".

Examples

# NOT RUN {
# calculate power
power_Beta(mu1 = 0.5, mu2 = 0.80, sd1 = 0.25, n1 = 60)
# calculate sample size for both groups
power_Beta(mu1 = 0.5, mu2 = 0.80, sd1 = 0.25, power=0.8)
# }