I have a question regarding survival analysis with exclusively censored data I have a set of 30 observations of length of alcohol abstinence ranging from 1 month to 120 months
The issue with the study's design is that length of alcohol abstinence was measured for patients without actual followup (patients with alcohol use disorder were simply asked for how long they have abstained from drinking alcohol before a liver transplant) Therefore, I know when the abstinence started but I do not when it might end in the future and no follow up is possible (therefore, all my data is right censored)
Interestingly enough, the distribution of these 30 observed length is perfectly log-normal
My question is, is it possible to fit a survival parametric model with a log-normal distribution of the errors even if all my observations are right censored ?
Also, what other options do I have for parametric regression modelling ? (coding in R)
Thanks kindly
