setPower: Representing the power

In randomizeR: Randomization for Clinical Trials

Representing the power

Description

Represents the expected power of the individual randomization sequences.

Usage

setPower(d, method, alpha = 0.05)

Arguments

d	effect size.
method	character string, should be one of "sim" or "exact", see Description.
alpha	significance level.

Details

The attained power of an individual randomization sequence can be an issue in the design of a clinical trial. The power of a randomization sequence is computed depending on the effect size d and the difference in group sizes.

If method="sim", the object represents the simulated power of an individual randomization sequence. When calling assess for a power object with method="sim", one test decision is computed for each randomization sequence of randSeq. The power is the proportion of falsely (correctly) rejected null hypotheses.

If method="exact", the object represents the exact power of an individual randomization sequence. When calling assess for a power object with method="exact", the exact p-value of each randomization sequence is computed. So far, this is only supported for normal endpoints. Then the power is the sum of the corresponding quantiles of the noncentral t-distribution.

Value

S4 object of class power, a formal representation of the issue of power in a clinical trial.

See Also

Other issues: chronBias, combineBias(), corGuess, imbal, issue, selBias

randomizeR documentation built on Sept. 19, 2023, 1:08 a.m.