Hierarchical Modeling and Prior Information in Toxicology

Explore a comprehensive Bayesian approach for estimating parameters in physiological pharmacokinetic models in this 1 hour 20 minute lecture by Andrew Gelman. Delve into the challenges of parameter estimation in complex models with numerous variables and limited observations. Learn about hierarchical population modeling for partial information pooling among subjects, a multi-compartment pharmacokinetic model, and the use of informative prior distributions for population parameters. Discover techniques for model estimation using Bayesian posterior simulation, which accounts for uncertainty in parameter-rich scenarios. Gain insights into model fit assessment and sensitivity analysis through posterior predictive simulation. This talk, presented at the Center for Language & Speech Processing (CLSP) at Johns Hopkins University, offers valuable knowledge for researchers and practitioners in toxicology, pharmacology, and related fields dealing with complex physiological models.