Zero-inflated latent growth modeling PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Resmi Gupta posted on Thursday, September 22, 2011 - 1:03 pm
I am modeling a latent variable (Y) which is measured on the basis of 5 (X1...X5) count items. Each item is measured over 6 time points. These items have excess number of zeroes and the proportion of zeros increases with time. Can mplus handle this type growth modeling ? Also, I was wondering if there is any article written by Drs. Muthen on this topic ? Thanks.
Reshmi Gupta
 Linda K. Muthen posted on Friday, September 23, 2011 - 9:48 am
Yes, this can be handled in Mplus. See the following paper which is available on the website:

Muthén, B. & Asparouhov, T. (2009). Growth mixture modeling: Analysis with non-Gaussian random effects. In Fitzmaurice, G., Davidian, M., Verbeke, G. & Molenberghs, G. (eds.), Longitudinal Data Analysis, pp. 143-165. Boca Raton: Chapman & Hall/CRC Press.
 Agnes Szabo posted on Thursday, February 01, 2018 - 5:04 am

I'm running a piecewise GMM with a count outcome using a zero-inflated Poisson model. The fit indices suggest a 5-class solution; however, the estimates for the second linear and quadratic slopes in one of the classes are 999. The same class emerges in the 4-class solution and the slopes are estimated fine in that model. Does this mean I should retain the 4-class solution?

Thank you for your response in advance!
 Bengt O. Muthen posted on Thursday, February 01, 2018 - 2:30 pm
That depends. Send the output for the 5-class solution to Support along with your license number.
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