This paper proposes approaches for the analysis of multiple change-
point models when dependency in the data is modelled through a hierarchical
Gaussian Markov random ¯eld. Integrated nested Laplace approximations are
used to approximate data quantities, and an approximate ¯ltering recursions ap-
proach is proposed for savings in compuational cost when detecting changepoints.
All of these methods are simulation free. Analysis of real data demonstrates the
usefulness of the approach in general. The new models which allow for data de-
pendence are compared with conventional models where data within segments is
assumed independent