Launch model, individually-varying ti... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Eric Thibodeau posted on Thursday, May 17, 2018 - 10:41 am
I'm working with a longitudinal study of marijuana use in the past 30 days. Big zero-inflated issue as you can imagine. I want to run a onset (survival) followed by growth (launch model). The data has 5 measurement occasions. Individual ages vary at measurement occasion. Is it possible to run a model with consideration of individually-varying times of observation? Can I bin people into age groups for the onset/survival part of the model, and then have my growth portion of the model tscores and AT command to set the five factor loadings to age at measurement occasion? IF so, would this require that my survival model be binned in five age groups? Thanks
 Tihomir Asparouhov posted on Thursday, May 17, 2018 - 7:57 pm
You can do all that but I would recommend instead to use tscores as in User's guide example 6.12, maybe after you standardize the time so that onset is time zero.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message