| Title: | PTE and RP for Optimally-Transformed Surrogate |
|---|---|
| Description: | Evaluates the strength of a surrogate marker by estimating the proportion of treatment effect explained (PTE) and relative power(RP) for the optimally-transformed version of the surrogate. Details available in Wang et al (2022) <arXiv:2209.08414>. |
| Authors: | Layla Parast [cre], Xuan Wang [aut] |
| Maintainer: | Layla Parast <[email protected]> |
| License: | GPL |
| Version: | 1.0 |
| Built: | 2025-02-10 02:42:27 UTC |
| Source: | https://github.com/cran/PTERP |
Hypothetical data for example
data("exampledata")data("exampledata")
the primary outcome
the surrogate marker
the treatment group indicator where 1 is treatment and 0 is control
data(exampledata)data(exampledata)
Evaluates the strength of a surrogate marker by estimating the proportion of treatment effect explained (PTE) and relative power(RP) for the optimally-transformed version of the surrogate.
PTERP(data,ncut=c(50,100,150,200,500,1000),n.resam=500)PTERP(data,ncut=c(50,100,150,200,500,1000),n.resam=500)
data |
data |
ncut |
relative power is calculated at a specific sample size n; this is a numeric vector of sample sizes for which the user wants the relative power, default is c(50,100,150,200,500,1000) |
n.resam |
number of times to resample, default is 500 |
A list of:
ptenew.es |
Proportion of treatment effect estimate |
rp_i1 |
Estimate of relative power at n=i1 where i1 is the first value given in ncut |
rp_i2 |
Estimate of relative power at n=i2 where i2 is the second value given in ncut, etc. |
pte.se |
Standard error estimate for the proportion of treatment effect explained |
rp.se.i1 |
Standard error estimate for the relative power at n=i1 where i1 is the first value given in ncut |
rp.se.i2 |
Standard error estimate for the relative power at n=i2 where i2 is the first value given in ncut, etc. |
Xuan Wang
data(exampledata) output=PTERP(exampledata,ncut=c(50,100,150,200,500,1000)) #reduce resampling output=PTERP(exampledata,ncut=c(50,100,150,200,500,1000),n.resam=5)data(exampledata) output=PTERP(exampledata,ncut=c(50,100,150,200,500,1000)) #reduce resampling output=PTERP(exampledata,ncut=c(50,100,150,200,500,1000),n.resam=5)