I'm attempting to do a continuous-time survival model where there is actually no censoring or truncation. I know this is a rare situation, but it causes a problem because Mplus won't let me import a variable with no variance and the continuous-time models won't run without the timecensored command. I'm new to Mplus, so perhaps there is an easy fix that I'm missing. Any help would be appreciated.
Boliang Guo posted on Tuesday, April 10, 2018 - 4:25 am
also, just noticed discrete time survival code not workable for data with a time point having no event. see example 6.19 with u0=0 which is true in reality as all subject are included but no event+ no censor at all.
DATA: FILE IS ex6.19.dat; VARIABLE: NAMES ARE u1-u4 x; CATEGORICAL = u1-u4 u0; MISSING = ALL (999); usevariable =u1-u4 u0; define: u0=0; ANALYSIS: ESTIMATOR = MLR; MODEL: f BY u0-u4@1; f ON x; f@0;
*** ERROR Categorical variable U0 contains less than 2 categories.