I'm trying to plot parametric survival curves for lognormal, exponential,log logistic and weibull distributions using AFT models.
Following is the toy example:
library(survival)
library(ISwR)
require(flexsurv)
ln.curves <- flexsurvreg(Surv(days, status == 1)~1,dist='lnorm',data=melanom)
ln.curves
wb.curves <- flexsurvreg(Surv(days, status == 1)~1,dist='weibull',data=melanom)
wb.curves
exp.curves <- flexsurvreg(Surv(days, status == 1)~1,dist='exponential',data=melanom)
exp.curves
ll.curves <- flexsurvreg(Surv(days, status == 1)~1,dist='llogis',data=melanom)
ll.curves
How would I pragmatically create survival function ? for example I would like to create a survival function for the following time sequenct.
t <- seq(1:5000)
I attempted to do this for log normal distribution, The parameters for log normal curves are:
Estimates:
est L95% U95% se
meanlog 8.670 8.302 9.038 0.188
sdlog 1.483 1.201 1.830 0.159
t <- seq(1:5000)
lnorm.plot <- 1 - plnorm(t, meanlog = 8.670,sdlog=1.483)
plot(lnorm.plot)
Below are my questions:
- Is my log normal curve plot correct?
- How to plot the same for exponential, weibull and loglogistic distribution?
