Missing data in discrete time surviva... PreviousNext
Mplus Discussion > Missing Data Modeling >
Message/Author
 Wei-Chen Chen posted on Monday, October 26, 2015 - 6:51 pm
Dear Mplus team:
I am running a simple model where two variables predict the survival function and then the survival function predict a distal outcome. My syntax was:
f by date0 date1@1 date2@1 date3@1 date4@1;
f on x1 x2;
x3 on f;
f@0;

Following examples from user guide, censoring was coded missing (including attrition and those who are out of risk set) in Mplus. However, what to do with the attrition of the distal outcome (wave 8)? The date0- date4 were from wave 1 to 5. Also, the covariance converge was very low between x2 and date2-4 as well as the distal outcome. Can I use weight to adjust for attrition in addition to Mplus' default? Thank you.
 Bengt O. Muthen posted on Tuesday, October 27, 2015 - 2:51 pm
I don't think there is much that can be done with strong attrition, except try to investigate how selective the attrition is, who is likely to fall out of the sample. With strong attrition, missing data methods rely too much on model assumptions.
 Wei-Chen Chen posted on Tuesday, October 27, 2015 - 5:38 pm
So, that means my survival model is also in big trouble because attrition is treated as censored right?
 Bengt O. Muthen posted on Tuesday, October 27, 2015 - 5:51 pm
Could be.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: