acondpow: Compute Approximate Conditional Power

View source: R/acondpow.R

acondpowR Documentation

Compute Approximate Conditional Power

Description

Computes an approximation to the conditional power of a sequential test based on the asymptotic normality of the statistic.

Usage

acondpow(za, power, zs, infr)

Arguments

za

standard normal critical value at final analysis

power

the overall planned power of the sequential test

zs

value of the test statistic on the standard normal scale at the current interim analysis

infr

information fraction (proportion of total information) at the current interim analysis

Details

Computes an approximation to the conditional probability of the test being rejected at the final analysis (i.e., that the test statistic will be > za), conditional on the current data, if the alternative for which the sequential test has power power is true. The approximation is based on asymptotic joint normality of the test statistic at the current and final analysis, and on the increment in the information being independent of the current data.

Value

The approximate conditional power (a scalar).

See Also

condpow; condpowcure

Examples

acondpow(2.0, 0.85, 0.32, 0.41)


raredd/desmon documentation built on May 7, 2024, 3:46 p.m.