plot.glmnetsurvfit: Plot survival and cumulative hazard curves

In CYGUBICKO/glmnetpostsurv: Fit Survival Models with 'glmnet'

Plot survival and cumulative hazard curves

Description

Plot estimated survival and cumulative hazard curves for glmnet model. glmnetsurvfit.glmnetsurv

Usage

## S3 method for class 'glmnetsurvfit'
plot(x, ..., type = c("surv", "cumhaz"), lsize = 0.3, compare = FALSE)

Arguments

x	a glmnetsurvfit.glmnetsurv or glmnetbasehaz.glmnetsurv object.
...	for future implementations
type	type of curve to generate. Either type = "surv" for survival curves or type = "cumhaz" for cumulative hazard curve.
lsize	line size for the curves. Default is 0.3.
compare	logical. Whether to return plot with labels to add additional geom object for comparison. Default is FALSE.

Details

Depending on the specification in glmnetsurvfit.glmnetsurv, this function plots either average or individual survival or cumulative hazard curves. The plot is a ggplot object, hence can be be customized further, see example below.

Value

a ggplot object.

Examples

library(ggplot2)
data(veteran, package="survival")

gmodel <- glmnetsurv(Surv(time, status) ~ factor(trt) + karno + diagtime + age + prior
	, data = veteran
	, lambda = 0
	, alpha = 1
)

# Survival estimate
gsurv <- glmnetsurvfit(fit = gmodel)

# Plot survival curves
plot(gsurv)

# Baseline survival estimate
gbsurv <- glmnetbasehaz(gmodel, centered = FALSE)
plot(gbsurv)

# Compare overall and baseline cumulative hazard
p1 <- plot(gsurv, type = "cumhaz", compare = TRUE)
df2 <- data.frame(time = gbsurv$time, cumhaz = gbsurv$hazard)
p2 <- (p1
	+ geom_line(data = df2, aes(x = time, y = cumhaz, group = 1, col = "baseline"))
	+ scale_colour_manual(name = "C. hazard"
		, values = c("#E41A1C", "#000000")
		, labels = c("baseline", "overall")
	)
)
print(p2)

CYGUBICKO/glmnetpostsurv documentation built on Sept. 1, 2022, 7:26 p.m.