Research Expertise and Interest

biostatistics, educational statistics, latent variable models, missing data methods, multilevel models, generalized linear latent and mixed models, hierarchical models, longitudinal data, Item response models, structural equation models

Research Description

Sophia Rabe-Hesketh is an applied statistician. She has developed a modeling framework called GLLAMM (Generalized Linear Latent and Mixed Modeling) for multilevel and latent variable modeling and written a publicly available software package called gllamm (http://www.gllamm.org/) to estimate these models. The theory of these models is published in the book Generalized Latent Variable Modeling, co-authored with Anders Skrondal. She has developed approximate methods for maximum likelihood estimation of models with high-dimensional latent variables and is currently doing research on Bayesian estimation, model evaluation, and missing data. Her papers are published in Psychometrika, Journal of Educational and Behavioral Statistics, Journal of Econometrics, Biometrics, and Journal of the Royal Statistical Society, Series A, among others. Sophia Rabe-Hesketh is also a member of the Interdepartmental Group in Biostatistics.