Dear Statalisters,
I'm using Stata 15.1.
Talking about survival analysis: I was asked to produce a Kaplan-Meier curve for my data, that are however a mixture of uncensored, right-censored, left-censored and interval-censored data. Given I have used two time variables in my analysis (a lower and an upper bound for time to event), it seems to me I cannot "stset" my data, thus I cannot produce a Kaplan-Meier nor a Neelson-Aalen function graph. Am I correct?
In fact, I have performed a stintreg regression, followed by stcurve.
Are there alternatives, to get a graph based on nonparametric analysis, to simply exclude left-censored and interval-censored data from the analysis, and then stset my data (since they would all be either uncensored or right-censored)?
Federico
I'm using Stata 15.1.
Talking about survival analysis: I was asked to produce a Kaplan-Meier curve for my data, that are however a mixture of uncensored, right-censored, left-censored and interval-censored data. Given I have used two time variables in my analysis (a lower and an upper bound for time to event), it seems to me I cannot "stset" my data, thus I cannot produce a Kaplan-Meier nor a Neelson-Aalen function graph. Am I correct?
In fact, I have performed a stintreg regression, followed by stcurve.
Are there alternatives, to get a graph based on nonparametric analysis, to simply exclude left-censored and interval-censored data from the analysis, and then stset my data (since they would all be either uncensored or right-censored)?
Federico
