A1099
Title: INLA: A computational tool for climate modeling
Authors: Janet Van Niekerk - King Abdullah University of Science and Technology (Saudi Arabia) [presenting]
Haavard Rue - KAUST (Saudi Arabia)
Abstract: Climate modeling often involves large data and/or complex models. In this realm, Bayesian modeling offers great promise to explain complex dependencies, but the computational complexity involved can deter the implementation of such models. Recent advances in the INLA methodology employ a Variational Bayes correction instead of the nested Laplace approximations, which results in INLA performing inference even faster and for more complex models or huge data. The purpose is to briefly introduce the modern INLA methodology and illustrate some examples of global climate models.
