Test for non-inferiority/superiority in Cox PH model.

Share:

Description

b is the log hazard ratio for treatment, δ is the margin

H0: b ≤ δ

Ha: b > δ

Usage

1
Cox.NIS(alpha, beta, loghr, p1, p2, d, margin)

Arguments

alpha

significance level

beta

power = 1-beta

loghr

log hazard ratio=log(lamda2/lamda1)=b

p1

the proportion of patients in treatment 1 group

p2

the proportion of patients in treatment 2 group

d

the probability of observing an event

margin

margin is the true difference of log hazard rates between control group lamda1 and a test drug group lamda2

References

Chow SC, Shao J, Wang H. Sample Size Calculation in Clinical Research. New York: Marcel Dekker, 2003

Examples

1
2
Example.7.3.4<-Cox.NIS(0.05,0.2,log(2),0.5,0.5,0.8,0.5)
Example.7.3.4

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.