BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:ABC methods for Bayesian model choice - Marin\, J-M (Universit Mon
 tpellier 2)
DTSTART:20140423T081500Z
DTEND:20140423T091500Z
UID:TALK52132@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Approximate Bayesian computation (ABC)\, also known as likelih
 ood-free methods\, have become a standard tool for the analysis of complex
  models\, primarily in population genetics. The development of new ABC met
 hodology is undergoing a rapid increase in the past years\, as shown by mu
 ltiple publications\, conferences and even software. While one valid inter
 pretation of ABC based estimation is connected with nonparametrics\, the s
 etting is quite different for model choice issues. We examined in Grelaud 
 et al. (2009\, Bayesian Analysis) the use of ABC for Bayesian model choice
  in the specific of Gaussian random fields (GRF)\, relying on a sufficient
  property only enjoyed by GRFs to show that the approach was legitimate. D
 espite having previously suggested the use of ABC for model choice in a wi
 der range of models in the DIYABC software (Cornuet et al.\, 2008\, Bioinf
 ormatics)\, we present in Robert et al. (2011\, PNAS) evidence that the ge
 neral use of ABC for model choice can be a real problem. Finally\, in Mari
 n et al. (2014\, JRSS B)\, we derive necessary and sufficient conditions o
 n summary statistics for the corresponding Bayes factor to be convergent\,
  namely to asymptotically select the true model. In this talk\, we will pr
 esent these different results. \n\nMarin\, Pillai\, Robert and Rousseau (2
 014) Relevant statistics for Bayesian model choice\, to appear in the Jour
 nal of the Royal Statistical Society\, Series B \n\nRobert\, Cornuet\, Mar
 in and Pillai (2011) Lack of confidence in approximate Bayesian computatio
 n model choice\, Proceedings of the National Academy of Science\, 108(37)\
 , 15112-15117 \n\nGrelaud\, Robert\, Marin\, Rodolphe and Taly (2009) ABC 
 likelihood-free methods for model choice in Gibbs random fields\, Bayesian
  Analysis\, 4(2)\, 317-336 \n\nCornuet\, Santos\, Beaumont\, Robert\, Mari
 n\, Balding\, Guillemaud and Estoup (2008) Inferring population history wi
 th DIY ABC: a user-friendly approach Approximate Bayesian Computation\, Bi
 oinformatics\, 24(23)\, 2713-2719\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR