Post date: Nov 23, 2012 4:30:34 PM
Course on Bayesian inference for latent Gaussian models using INLA
On January 24, 2013, a one day course on Bayesian inference for latent Gaussian models using INLA will be organized at the National Institute for Public Health and the Environment – RIVM. The course will be presented by Håvard Rue of the Department of Mathematical Sciences, Norwegian University of Science and Technology, Trondheim, Norway.
Course information
INLA (Integrated Nested Laplace Approximations) revolutionizes the computer performance of Bayesian inference for latent Gaussian models. Using this recent tool, very accurate approximations to the posterior marginals can be computed directly. The main benefit of these approximations is computational: where MCMC algorithms need hours or days to run, INLA provides precise estimates in seconds or minutes. Another advantage is its generality, which makes it possible to perform Bayesian analysis in an automatic, streamlined way.
Latent Gaussian models (LGMs) are perhaps the most commonly used class of models in statistical modeling applications. It includes, among others:
- (generalized) linear mixed models
- (generalized) additive models
- smoothing spline models and semi-parametric regression models
- spatial models
- temporal models
- log-Gaussian Cox processes
- geostatistical models
Possible applications are found in:
- longitudinal data analysis
- time series analysis
- disease mapping
- spatial survival analysis
- spatial association studies
- point processes
- interpolation of spatial data
This one day course will provide an overview of the INLA approach and possible (also spatial) applications.
The course is intended for statisticians, epidemiologists and other researchers who want to model their data using LGMs. Basic knowledge of Bayesian statistics and generalized linear and additive models is recommended.
More information on INLA can be found athttp://www.r-inla.org/
References:
H. Rue, S. Martino, and N. Chopin. Approximate Bayesian inference for latent Gaussian models using integrated nested Laplace approximations (with discussion). Journal of the Royal Statistical Society, Series B,71(2):319{392, 2009
F. Lindgren, H. Rue, and J. Lindstrom. An explicit link between Gaussian fields and Gaussian Markov random fields: The SPDE approach (with discussion). Journal of the Royal Statistical Society, Series B, 73(4):423–498, 2011.
Practical information
Organizers
Jan van de Kassteele & Caroline Ameling
National Institute for Public Health and the Environment – RIVM
Date and time
January 24, 2013. 9.00 – 17.00h.
Location
National Institute for Public Health and the Environment – RIVM, Bilthoven, the Netherlands.
Fee
50 euro. Registration is required. The fee covers lunch and local costs to arrange the course. Payment by invoice.
Registration
Registration is done by e-mailing secretary Anne-Marie van Kleef, Anne.Marie.van.Kleef@rivm.nl. Deadline for registration is January 18, 2013.
Jan van de Kassteele, PhD
Biostatistician
Department of Statistics, Mathematical Modelling and Data Logistics
National Institute for Public Health and the Environment - RIVM
P.O. Box 1 | 3720 BA Bilthoven | The Netherlands | tel.: +31-30-2743690 | fax.: +31-30-2744456 | e-mail: jan.van.de.kassteele@rivm.nl