J. Sunil Rao, of the Division of Biostatistics at the University of Miami, will present:
“A Tour of Classified Mixed Model Predictions and Projections”
Abstract: Many practical problems are related to prediction where the main interest is at the subject or small sub-population level. In such cases, it’s possible to make substantial gains in prediction accuracy by identifying a class that a new subject belongs to and associating the new subject with a random effect corresponding to the same class in the training data so that mixed model prediction can be used. In this talk, we first introduce the original classified mixed model prediction idea and then discuss some newer developments in multivariate classified mixed model prediction and classified mixed model projections for new data outside the range of the training data. This is joint work over the years with Jiming Jiang (UC-Davis), Thuan Nguyen (OHSU) and former UM students Menying Li (Moderna), Hang Zhang (Biogen) and Jie Fan (Novartis).
All are Welcome.