mets (version 1.3.4)

logitSurv: 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 coxph)

data

data frame

offset

offsets for exp(x beta) terms

weights

weights for score equations

...

Additional arguments to lower level funtions

Author

Thomas Scheike

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.

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)