As compared to other methods of survival analysis, discrete time survival analysis analyzes time in discrete chunks during which the event of interest could occur. For example, suppose you were studying dropping out of school but only knew the grade in which someone dropped out (e.g., 10th grade). You could analyze "time to dropout" using discrete time survival analysis. This seminar will explore such methods, but also extend them further exploring how, when using a latent variable framework, specifically latent class regression, unobserved heterogeneity in the population with regards to the event process and susceptibility to risk factors can be incorporated into the model. Using this framework, multivariate forms of survival data can also be accommodated. Analyses will be illustrated using Mplus version 3.
You can view an extended version of what was presented in this seminar, courtesy of Dr. Masyn's presentation in Professor Muthén's Ed 231E Course
The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California.