Bayesian statistics: An introduction to multilevel models and the R package MCMCglmm


07-Oct-2009 - 09-Oct-2009

Bayesian statistics become increasingly important in the field of evolutionary biology and particularly for analyses involving multilevel and/or hierarchical models such as those used in comparative studies and in animal models. The aim of this workshop is to give an introduction to the use of Bayesian statistics as an alternative to other statistical approaches for hierarchical modeling (e.g. maximum likelihood). The statistical package MCMCglmm for R was created by Dr Jarrod Hadfield and Dr Loeske Kruuk and published at the beginning of 2009. This package is a useful and powerful tool based on Bayesian statistics for all scientists interested in multi-level/hierarchical models applied in many different fields of evolutionary biology. During this 3-day workshop, Dr Hadfield will introduce the participants to (1) the basics of glmm and MCMC, (2) multivariate and random regressions and (3) the use of MCMCglmm with pedigrees and phylogenies. Participants need to have a basic knowledge of statistics and Bayesian methods and of the programming language R. The course is free and is be open to a maximum of 25 participants selected by order of registration.


Teacher: Dr Jarrod Hadfield


Venue: Evolutionary Biology Centre


Form: 3-day workshop


Registration: Contact Simone Immler simone.immler@ebc.uu.se) to register for this course.