search

drizopoulos / JM

Joint Models for Longitudinal & Survival Data under Maximum Likelihood
34 stars 7 forks source link

problem with scaling of time variables #16

Closed CaryOberije closed 3 years ago

CaryOberije commented 4 years ago

Dear Dimitris, I'm trying to use the JM library, but I get error messages and they seem connected to the time duration units: days or months.

I was using months as time scale and I then got the message: "Error in aeqSurv(Y) : aeqSurv exception, an interval has effective length 0" It seems that this happens if I use months for survival analysis.

If I change my time variables into duration in days, I get another error message: "Error in if (t1 || t2) { : missing value where TRUE/FALSE needed" I found a similar issue here in the GitHub and the advice was not to use days as unit for duration.

I'm a bit stuck with this as nor the days neither the months seem to work. I would very much appreciate it if you could give me some advice?

drizopoulos commented 3 years ago

It's difficult to see what is going on without a reproducible example.