cif2.logrank {surv2sample} | R Documentation |
Compares cumulative incidence functions (CIF) for one failure cause in two samples of censored competing risks data using Gray's G^rho-weighted logrank-type test based on subdistribution hazards corresponding to CIFs.
cif2.logrank(x, group, cause = 1, rho = 0)
x |
a "Survcomp" object, as returned by the
Survcomp function. |
group |
a vector indicating to which group each observation belongs. May contain values 1 and 2 only. |
cause |
For which cause of failure should the CIFs be compared? |
rho |
the parameter (exponent) of the weight. |
The test compares cumulative incidence functions F_1(t,k), F_2(t,k) for a particular failure cause k.
This test was proposed by Gray (1988). The statistic is similar to the weighted logrank test, but instead of ordinary hazards this test compares subdistribution hazards, which is equivalent to comparing CIFs (due to the one-to-one relationship between subdistribution hazards and CIFs). The G^rho weight is of the form (1-F_0(t,k))^rho, where F_0(t,k) is a pooled sample estimate of the cause k cumulative incidence curve.
This logrank-type test compares subdistribution hazards, and should not be confused with the ordinary logrank test applied to comparing cause-specific hazards. Hypotheses on subdistribution hazards (or CIFs) and cause-specific hazards are not equivalent.
A list of class "cif2.logrank"
with components:
stat |
the test statistic. |
pval |
the p-value based on the asymptotic normality. |
Further components are cause
and rho
, the same
as on input.
David Kraus (http://www.davidkraus.net/)
Gray, R. J. (1988) A class of k-sample tests for comparing the cumulative incidence of a competing risk. Ann. Statist. 16, 1141–1154.
cif
and plot.cif
for estimation and
plotting of CIFs, cif2.ks
, cif2.int
and
cif2.neyman
for other two-sample tests.
## bone marrow transplant data data(bmt1) ## compare CIFs for cause 1 (relapse) cif2.logrank(Survcomp(bmt1$time, bmt1$event), bmt1$donor, cause = 1) ## compare CIFs for cause 2 (death in remission) cif2.logrank(Survcomp(bmt1$time, bmt1$event), bmt1$donor, cause = 2)