logitSurv {mets} | R Documentation |
Proportional odds survival model
Description
Semiparametric Proportional odds model, that has the advantage that
logit(S(t|x)) = \log(\Lambda(t)) + x \beta
so covariate effects give OR of survival.
Usage
logitSurv(formula, data, offset = NULL, weights = NULL, ...)
Arguments
formula |
formula with 'Surv' outcome (see |
data |
data frame |
offset |
offsets for exp(x beta) terms |
weights |
weights for score equations |
... |
Additional arguments to lower level funtions |
Details
This is equivalent to using a hazards model
Z \lambda(t) \exp(x \beta)
where Z is gamma distributed with mean and variance 1.
Author(s)
Thomas Scheike
References
The proportional odds cumulative incidence model for competing risks, Eriksson, Frank and Li, Jianing and Scheike, Thomas and Zhang, Mei-Jie, Biometrics, 2015, 3, 687–695, 71,
Examples
data(TRACE)
dcut(TRACE) <- ~.
out1 <- logitSurv(Surv(time,status==9)~vf+chf+strata(wmicat.4),data=TRACE)
summary(out1)
gof(out1)
plot(out1)
[Package mets version 1.3.4 Index]