skip to content

Scalable statistical inference with INLA

Presented by: 
Havard Rue
Monday 3rd July 2017 - 16:15 to 17:00
INI Seminar Room 1
INLA do approximate Bayesian inference for the class of latent Gaussian models. It has shown sucessful allowing statisticians and applied scientists to fast and reliable Bayesian inference for a huge class of additve models, within reasonable time. Especially, the use of spatial Gaussian models using the SPDE approach has been particularly popular. Although most models runs within reasonable time, we are facing with the current implementation, limitations for really huge models like large space time models. In this talk I will discuss the current situation and possible strategies to improve the situation.
The video for this talk should appear here if JavaScript is enabled.
If it doesn't, something may have gone wrong with our embedded player.
We'll get it fixed as soon as possible.
University of Cambridge Research Councils UK
    Clay Mathematics Institute London Mathematical Society NM Rothschild and Sons