Cumulative residuals after model matrix for Cox PH regression p-values based on Lin, Wei, Ying resampling.
gofM.phreg(
formula,
data,
offset = NULL,
weights = NULL,
modelmatrix = NULL,
n.sim = 1000,
silent = 1,
...
)formula for cox regression
data for model
offset
weights
matrix for cumulating residuals
number of simulations for score processes
to keep it absolutely silent, otherwise timing estimate will be prduced for longer jobs.
Additional arguments to lower level funtions
That is, computes $$ U(t) = \int_0^t M^t d \hat M $$ and resamples its asymptotic distribution.
This will show if the residuals are consistent with the model. Typically, M will be a design matrix for the continous covariates that gives for example the quartiles, and then the plot will show if for the different quartiles of the covariate the risk prediction is consistent over time (time x covariate interaction).
library(mets)
data(TRACE)
set.seed(1)
TRACEsam <- blocksample(TRACE,idvar="id",replace=FALSE,100)
dcut(TRACEsam) <- ~.
mm <- model.matrix(~-1+factor(wmicat.4),data=TRACEsam)
m1 <- gofM.phreg(Surv(time,status==9)~vf+chf+wmi,data=TRACEsam,modelmatrix=mm)
summary(m1)
#> Cumulative residuals versus modelmatrix :
#> Sup_t |U(t)| pval
#> factor(wmicat.4)[0.4,1.1] 5.788752 0.021
#> factor(wmicat.4)(1.1,1.4] 2.633143 0.591
#> factor(wmicat.4)(1.4,1.72] 5.657370 0.038
#> factor(wmicat.4)(1.72,2] 1.227485 0.948
#>
#> Cumulative score process versus covariates (discrete z via model.matrix):
#> Sup_z |U(tau,z)| pval
#> matrixZ 1.801626 0.812
if (interactive()) {
par(mfrow=c(2,2))
plot(m1)
}
m1 <- gofM.phreg(Surv(time,status==9)~strata(vf)+chf+wmi,data=TRACEsam,modelmatrix=mm)
summary(m1)
#> Cumulative residuals versus modelmatrix :
#> Sup_t |U(t)| pval
#> factor(wmicat.4)[0.4,1.1] 6.049754 0.022
#> factor(wmicat.4)(1.1,1.4] 2.211031 0.757
#> factor(wmicat.4)(1.4,1.72] 5.040704 0.085
#> factor(wmicat.4)(1.72,2] 1.482109 0.890
#>
#> Cumulative score process versus covariates (discrete z via model.matrix):
#> Sup_z |U(tau,z)| pval
#> matrixZ 1.228032 0.934
## cumulative sums in covariates, via design matrix mm
mm <- cumContr(TRACEsam$wmi,breaks=10,equi=TRUE)
m1 <- gofM.phreg(Surv(time,status==9)~strata(vf)+chf+wmi,data=TRACEsam,
modelmatrix=mm,silent=0)
#> Cumulative score process test for modelmatrix:
#> Sup_t |U(t)| pval
#> <=0.56 0.84 0.35
#> <=0.72 1.93 0.39
#> <=0.88 5.06 0.04
#> <=1.04 2.63 0.58
#> <=1.2 4.27 0.17
#> <=1.36 4.04 0.17
#> <=1.52 3.20 0.29
#> <=1.68 1.83 0.81
#> <=1.84 1.25 0.82
#> <=2 0.00 1.00
summary(m1)
#> Cumulative residuals versus modelmatrix :
#> Sup_t |U(t)| pval
#> <=0.56 0.8423856 0.352
#> <=0.72 1.9258405 0.394
#> <=0.88 5.0552596 0.037
#> <=1.04 2.6334412 0.579
#> <=1.2 4.2747875 0.174
#> <=1.36 4.0393945 0.169
#> <=1.52 3.2004455 0.292
#> <=1.68 1.8330704 0.810
#> <=1.84 1.2509429 0.823
#> <=2 0.0000000 1.000
#>
#> Cumulative score process versus covariates (discrete z via model.matrix):
#> Sup_z |U(tau,z)| pval
#> matrixZ 3.73439 0.283