Online Program Home
My Program

Abstract Details

Activity Number: 563 - Recent Advances in Bayesian Structure Learning
Type: Topic Contributed
Date/Time: Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #306513
Title: Bayesian Semiparametric Functional Mixed Models
Author(s): Abhra Sarkar* and Giorgio Paulon and Bharath Chandrsekaran and Fernando Llanos
Companies: The University of Texas at Austin and The University of Texas at Austin and University of Pittsburg and University of Pittsburgh
Keywords: Bayesian Statistics; Longitudinal Data; Learning Curves; Mixed Models
Abstract:

We consider the problem of flexible modelling of learning curves in longitudinal experiments. We propose Bayesian semiparametric functional mixed effects models that can characterize unrestricted as well as practically meaningful shape-restricted classes of learning curves while also allowing assessment of globar-local influences of exogenous predictors. We design Markov chain Monte Carlo algorithms for posterior computation. We illustrate the method's efficacy using numerical experiments. Applications to learning curves from speech categorization experiments illustrate the method's practical utility.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program