A pruned recursive solution to the multiple change point problem
- 193 Downloads
Long time series are often heterogeneous in nature. As such, the most appropriate model is one whose parameters are allowed to change through time. The exponential number of solutions to the multiple change point problem requires an efficient algorithm in order to be computationally feasible. Exact Bayesian solutions have at best quadratic complexity in the number of observations, which still can be too slow for very large data sets. Here, a pruned dynamic programming algorithm is proposed to fit a piecewise regression model with unknown break points to a data set. The algorithm removes unessential calculations, reducing the complexity of the most time consuming step of the algorithm from quadratic in the number of observations to quadratic in the average distance between change points. A distance measure is introduced that can be used to determine the divergence of the approximate joint posterior distribution from the exact posterior distribution. Analysis of two real data sets shows that this approximate algorithm produces a nearly identical representation of the joint posterior distribution on the locations of the change points, but with a significantly faster run time than its exact counterpart.
KeywordsBayesian change point Dynamic programming Joint posterior Kullback–Leibler divergence Piecewise regression
The author would like to thank the two anonymous reviewers for their thoughtful feedback which helped to greatly improve this manuscript. This work was supported by a grant from the National Science Foundation, DMS-1407670 (E. Ruggieri, PI).
- Adams RP, MacKay DJC (2007) Bayesian online changepoint detection. http://arxiv.org/pdf/0710.3742.pdf. Accessed 20 June 2016
- Fryzlewicz P (2013) Wild binary segmentation for multiple change-point detection. http://stats.lse.ac.uk/fryzlewicz/wbs/wbs.pdf. Accessed 20 June 2016
- Killick R, Nam CFH, Aston JAD, Eckley IA (2012b) Changepoint.info: the changepoint repository. http://changepoint.info
- Milankovitch M (1941) Canon of insolation and the ice-age problem. Israel program for scientific translations, Jerusalem (1969)Google Scholar
- Rigaill G (2010) Pruned dynamic programming for optimal multiple change-point detection. http://arXiv:1004.0887v2.pdf . Accessed 20 June 2016Google Scholar
- Ross GJ (2013) Parametric and nonparametric sequential change detection in R: the cpm package. http://www.gordonjross.co.uk/cpm.pdf. Accessed 20 June 2016
- Ruddiman WF (2013) Earth’s climate: past and future, 3rd edn. WH Freeman, New YorkGoogle Scholar
- Saatci Y, Turner R, Rasmussen CE (2010) Gaussian process change point models. In: Proceedings of the 27th international conference on machine learning, pp 927–934Google Scholar
- Whiteley N, Andrieu C, Doucet A (2011) Bayesian computational methods for inference in multiple change-point models. http://www.maths.bris.ac.uk/~manpw/change_points_2011.pdf. Accessed 20 June 2016