Questions on Group-Based Trajectory Modeling (GBTM) for Categorical Outcomes

General
1

I am working on a project involving group-based trajectory modeling (GBTM) and have several questions I hope you can help with. My dataset tracks treatment adherence (a binary variable: yes/no) over several years (2012-2016). Here are my queries:

  1. Can GBTM be applied to categorical outcomes?
  2. Which variables should be considered to identify trajectories? Should I use just the time variable (year) and the outcome (treatment adherence), or do I need to adjust for other covariates?
  3. What should I do if the odds of correct classification (OCC) for several groups is less than 5 and the average posterior probability is below 0.7?
  4. Why am I getting random BIC values for increasing degrees? Usually, the BIC decreases with increasing degrees, but in my case, for 1 group it might be -7800, for 2 groups it might be -7600, and for 3 groups it might be -7700.

I am a layman without much knowledge on statistics or R. Please, answer to the questions in digestible way for me. Thank you in advance for your assistance. Your insights will be greatly appreciated.

2

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.