Classifications, Profiles, and Latent Growth Mixture Models

Catalog description

Multivariate strategies for prediction of binary, ordered and categorical outcomes and for classification accuracy and hypothesis testing; finite mixture modeling for discovery, validity and explanation of latent subgroups of profiles and change trajectories. Methods include multiple logistic and multinomial logistic regression, Receiver Operating Characteristic curve analysis, multiple discriminant analysis, hierarchical cluster analysis, multidimensional classification analysis, latent profile analysis, latent transition analysis, and latent growth mixture modeling.