Bayesian multiple change-point estimation with annealing stochastic approximation Monte Carlo (Q2430237)

The aim of the paper is to set up a Bayesian multiple change-point model and obtain the posterior distribution of the positions of change-points when both the number of change-points and their locations are unknown. Bayesian multiple change-point problems for the normal, exponential, binomial and Poisson distributions are described and each posterior distribution of change-points is provided. As a computational tool for posterior probability calculation, the Annealing Stochastic Approximation Monte Carlo (ASAMC) algorithm is adopted for change-point identification. Some numerical results of the simulations and the comparison of ASAMC and Evolutionary Monte Carlo (EMC) are presented. The methodology is applied to temperature and the number of respiratory deaths in Seoul, South Korea, in 1999.