B1072
Title: Improved inference for LGM's using INLA
Authors: Janet Van Niekerk - King Abdullah University of Science and Technology (Saudi Arabia) [presenting]
Denis Rustand - KAUST (Saudi Arabia)
Elias Krainski - KAUST (Brazil)
Haavard Rue - KAUST (Saudi Arabia)
Abstract: INLA has gained popularity as an approximate inferential method due to its accuracy and efficiency for the class of latent Gaussian models. We present recent advancements in the methodology, implemented in the R-INLA library, that performs approximate Bayesian inference even faster while achieving the same accuracy. This advancement avails INLA as an inferential framework for huge data such as fMRI analysis and complex joint survival models.