Open syclik opened 9 years ago
From @bob-carpenter on December 4, 2014 2:52
I have to add right away that I'm not a huge fan of this kind of function that overloads y
as either a censoring bound or an outcome. I'd prefer
y_obs ~ weibull(alpha,sigma);
y_cens ~ weibull_ccdf(alpha,sigma);
with two different variables, y_obs
for the ordinary observations and y_cens
for right censored. If we need to do left censoring, it's just the CDF instead of the CCDF. If we need to do both, we need a new notation; see #1154
From @bob-carpenter on December 4, 2014 2:48
Aki Vehtari suggested building in a function that behaves like some R survival analysis packages:
The int value
z
is used as an indicator as to whether the event was right censored aty
or whether it was an ordinary outcome.Aki also mentioned that it's common to have a vector of censored outcomes, all censored at the same place.
Copied from original issue: stan-dev/stan#1153