CMStatistics 2015: Start Registration
View Submission - CMStatistics
B1039
Title: Calibration of conditional composite likelihood for Bayesian inference on Gibbs random fields Authors:  Nial Friel - University College Dublin (Ireland)
Julien Stoehr - University College Dublin (France) [presenting]
Abstract: Gibbs random fields play an important role in statistics, however the resulting likelihood is typically unavailable due to an intractable normalizing constant. Composite likelihoods offer a principled means to construct useful approximations. An approach will be presented to calibrate the posterior distribution resulting from using a composite likelihood and illustrate its performance in several examples.