Hi, John,
it should work the same way as without interaction (but make sure
to use the fitted object "coxfit", not just "fit" in our call of
survfit, and note that age*rx already expands to age + rx + age:rx
so that age + rx is redundant in your formula):
coxfit <- coxph(Surv(futime, fustat) ~ a
Perhaps this might be useful:
https://rpubs.com/tf_peterson/interactionplotDemo
Bert Gunter
"The trouble with having an open mind is that people keep coming along and
sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Thu, May 20, 2021 at 10:29 AM S
Colleagues,
I hope someone can tell me how to plot a cox model that contains an interaction
term.
I know that plot(survfit(. . . . )) can be used to plot a Cox model, i.e..
coxfit <- coxph(Surv(futime, fustat) ~ age+rx, data = ovarian)
plot(survfit(fit, newdata=data.frame(age=60,rx=2)))
but
3 matches
Mail list logo
