Adaptive Sequential Monte Carlo for Multiple Changepoint Analysis

ABSTRACT Process monitoring and control requires the detection of structural changes in a data stream in real time. This article introduces an efficient sequential Monte Carlo algorithm designed for learning unknown changepoints in continuous time. The method is intuitively simple: new changepoints for the latest window of data are proposed by conditioning only on data observed since the most recent estimated changepoint, as these observations carry most of the information about the current state of the process. The proposed method shows improved performance over the current state of the art. Another advantage of the proposed algorithm is that it can be made adaptive, varying the number of particles according to the apparent local complexity of the target changepoint probability distribution. This saves valuable computing time when changes in the changepoint distribution are negligible, and enables rebalancing of the importance weights of existing particles when a significant change in the target distribution is encountered. The plain and adaptive versions of the method are illustrated using the canonical continuous time changepoint problem of inferring the intensity of an inhomogeneous Poisson process, although the method is generally applicable to any changepoint problem. Performance is demonstrated using both conjugate and nonconjugate Bayesian models for the intensity. Appendices to the article are available online, illustrating the method on other models and applications.

[1]  D. Hand,et al.  Bayesian anomaly detection methods for social networks , 2010, 1011.1788.

[2]  Ajay Jasra,et al.  Inference for a class of partially observed point process models , 2012, 1201.4529.

[3]  Jun S. Liu,et al.  Metropolized independent sampling with comparisons to rejection sampling and importance sampling , 1996, Stat. Comput..

[4]  P. Fearnhead,et al.  On‐line inference for hidden Markov models via particle filters , 2003 .

[5]  A. Raftery,et al.  Bayesian analysis of a Poisson process with a change-point , 1986 .

[6]  G. Kitagawa.  Monte Carlo Filter and Smoother for Non-Gaussian Nonlinear State Space Models , 1996 .

[7]  P. Moral,et al.  Sequential Monte Carlo samplers , 2002, cond-mat/0212648.

[8]  Joseph Goedert.  Following a moving target. , 2008, Health data management.

[9]  Simon J. Godsill,et al.  Monte Carlo Filtering of Piecewise Deterministic Processes , 2011 .

[10]  P. Fearnhead,et al.  On‐line inference for multiple changepoint problems , 2007 .

[11]  Bin Wu,et al.  Cell phone mini challenge award: Social network accuracy— exploring temporal communication in mobile call graphs , 2008, 2008 IEEE Symposium on Visual Analytics Science and Technology.

[12]  Dieter Fox,et al.  Adapting the Sample Size in Particle Filters Through KLD-Sampling , 2003, Int. J. Robotics Res..

[13]  W. Gilks,et al.  Following a moving target—Monte Carlo inference for dynamic Bayesian models , 2001 .

[14]  N. Chopin.  Dynamic Detection of Change Points in Long Time Series , 2007 .

[15]  Nicholas A. Heard,et al.  Convergence of Monte Carlo distribution estimates from rival samplers , 2016, Stat. Comput..

[16]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .