×

MCMC implementation for Bayesian hidden semi-Markov models with illustrative applications. (English) Zbl 1322.62205

Summary: Hidden Markov models (HMMs) are flexible, well-established models useful in a diverse range of applications. However, one potential limitation of such models lies in their inability to explicitly structure the holding times of each hidden state. Hidden semi-Markov models (HSMMs) are more useful in the latter respect as they incorporate additional temporal structure by explicit modelling of the holding times. However, HSMMs have generally received less attention in the literature, mainly due to their intensive computational requirements. Here a Bayesian implementation of HSMMs is presented. Recursive algorithms are proposed in conjunction with Metropolis-Hastings in such a way as to avoid sampling from the distribution of the hidden state sequence in the MCMC sampler. This provides a computationally tractable estimation framework for HSMMs avoiding the limitations associated with the conventional EM algorithm regarding model flexibility. Performance of the proposed implementation is demonstrated through simulation experiments as well as an illustrative application relating to recurrent failures in a network of underground water pipes where random effects are also included into the HSMM to allow for pipe heterogeneity.

MSC:

62M05 Markov processes: estimation; hidden Markov models
62F15 Bayesian inference
60K15 Markov renewal processes, semi-Markov processes
60J22 Computational methods in Markov chains

Software:

R; hsmm; BayesDA
PDFBibTeX XMLCite
Full Text: DOI Link

References:

[1] R Development Core Team: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (2012)
[2] Baum, L.E., Petrie, T., Soules, G., Weiss, N.: A maximization technique in the statistical analysis of probabilistic functions of Markov chains. Ann. Math. Stat. 41, 164-171 (1970) · Zbl 0188.49603 · doi:10.1214/aoms/1177697196
[3] Bellone, E., Hughes, J.P., Guttorp, P.: A hidden Markov model for downscaling synoptic atmospheric patterns to precipitation amounts. Clim. Res. 15, 1-12 (2000) · doi:10.3354/cr015001
[4] Bulla, J., Bulla, I.: Stylized facts of financial time series and hidden semi-Markov models. Comput. Stat. Data Anal. 51, 2192-2209 (2006) · Zbl 1157.62518 · doi:10.1016/j.csda.2006.07.021
[5] Bulla, J., Bulla, I., Nenadic, O.: HSMM—an R package for analyzing hidden semi-Markov models. Comput. Stat. Data Anal. 54, 611-619 (2010) · Zbl 1464.62011 · doi:10.1016/j.csda.2008.08.025
[6] Celeux, G., Hurn, M., Robert, C.P.: Computational and inferential difficulties with mixture posterior distributions. J. Am. Stat. Assoc. 95(451), 957-970 (2000) · Zbl 0999.62020 · doi:10.1080/01621459.2000.10474285
[7] Chib, S.: Calculating posterior distributions and modal estimates in Markov mixture models. J. Econom. 75(1), 79-97 (1996) · Zbl 0864.62010 · doi:10.1016/0304-4076(95)01770-4
[8] Devijver, P.A.: Baum’s forward-backward algorithm revisited. Pattern Recognit. Lett. 3(6), 369-373 (1985) · Zbl 0593.62083 · doi:10.1016/0167-8655(85)90023-6
[9] Dewar, M., Wiggins, C., Wood, F.: Inference in hidden Markov models with explicit state duration distributions. IEEE Signal Process. Lett. 19(4), 235-238 (2012) · doi:10.1109/LSP.2012.2184795
[10] Dong, M., He, D.: A segmental hidden semi-Markov model (HSMM)-based diagnostics and prognostics framework and methodology. Mech. Syst. Signal Process. 21, 2248-2266 (2007) · doi:10.1016/j.ymssp.2006.10.001
[11] Economou, T.; Vitolo, R.; Bailey, T. C.; Kapelan, Z.; Waterhouse, E. K., A latent structure model for high river flows, 125-129 (2009)
[12] Economou, T., Kapelan, Z., Bailey, T.C.: On the prediction of underground water pipe failures: zero-inflation and pipe specific effects. J. Hydroinform. 14(4), 872-883 (2012) · doi:10.2166/hydro.2012.144
[13] Fearnhead, P., Sherlock, C.: An exact Gibbs sampler for the Markov-modulated Poisson process. J. R. Stat. Soc., Ser. B, Stat. Methodol. 68(5), 767-784 (2006) · Zbl 1110.62131 · doi:10.1111/j.1467-9868.2006.00566.x
[14] Ferguson, J. D.; Ferguson, J. D. (ed.), Variable duration models for speech, Princeton, NJ
[15] Gamerman, D.: Markov Chain Monte Carlo: Stochastic Simulation for Bayesian Inference. Chapman & Hall, London (1997) · Zbl 0881.62002
[16] Gelman, A., Roberts, G.O., Gilks, W.R.: Efficient Metropolis jumping rules. Bayesian Stat. 5, 599-607 (1996)
[17] Gelman, A., Carlin, J.B., Stern, H.S., Rubin, D.B.: Bayesian Data Analysis. Chapman & Hall/CRC, London (2004) · Zbl 1039.62018
[18] Gilks, W.R., Richardson, S., Spiegelhalter, D.J.: Markov Chain Monte Carlo in Practice. Chapman & Hall/CRC, London (1996) · Zbl 0832.00018 · doi:10.1007/978-1-4899-4485-6
[19] Guedon, Y.: Review of several stochastic speech unit models. Comput. Speech Lang. 6, 377-402 (1992) · doi:10.1016/0885-2308(92)90030-8
[20] Guedon, Y.: Estimating hidden semi-Markov chains from discrete sequences. J. Comput. Graph. Stat. 12(3), 604-639 (2003) · doi:10.1198/1061860032030
[21] Guha, S., Li, Y., Neuberg, D.: Bayesian hidden Markov modeling of array CGH data. J. Am. Stat. Assoc. 103(482), 485-497 (2008) · Zbl 1469.62368 · doi:10.1198/016214507000000923
[22] Hughes, J.P., Guttorp, P., Charles, S.P.: A non-homogeneous hidden Markov model for precipitation occurrence. J. R. Stat. Soc., Ser. C, Appl. Stat. 48(1), 15-30 (1999) · Zbl 0920.62141 · doi:10.1111/1467-9876.00136
[23] Jardine, A.K., Lin, D., Banjevic, D.: A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mech. Syst. Signal Process. 20(7), 1483-1510 (2006) · doi:10.1016/j.ymssp.2005.09.012
[24] Johnson, M.J., Willsky, A.S.: Bayesian nonparametric hidden semi-Markov models. arXiv:1203.1365v2 (2012) · Zbl 1320.62050
[25] Jouyaux, C., Richardson, S., Longini, I.: Modeling markers of disease progression by a hidden Markov process: application to characterizing cd4 cell decline. Biometrics 56(3), 733-741 (2000) · Zbl 1060.62619 · doi:10.1111/j.0006-341X.2000.00733.x
[26] Kleiner, Y., Rajani, B.: Comprehensive review of structural deterioration of water mains: statistical models. Urban Water 3, 131-150 (2001) · doi:10.1016/S1462-0758(01)00033-4
[27] Kozumi, H.: Bayesian analysis of discrete survival data with a hidden Markov chain. Biometrics 56(4), 1002-1006 (2000) · Zbl 1060.62509 · doi:10.1111/j.0006-341X.2000.01002.x
[28] Levinson, S.E.: Continuously variable duration hidden Markov models for automatic speech recognition. Comput. Speech Lang. 1, 29-45 (1986) · doi:10.1016/S0885-2308(86)80009-2
[29] Marin, J.-M., Robert, C.P.: Bayesian Core: A Practical Approach to Computational Bayesian Statistics. Springer, Berlin (1997)
[30] Rabiner, L.: A tutorial on hidden Markov models and selected applications in speech recognition. Proc. IEEE 77(2), 257-285 (1989) · doi:10.1109/5.18626
[31] Richardson, S., Green, P.J.: On Bayesian analysis of mixtures with an unknown number of components. J. R. Stat. Soc. B 59(4), 731-792 (1997) · Zbl 0891.62020 · doi:10.1111/1467-9868.00095
[32] Robert, C.P., Titterington, D.M.: Reparameterization strategies for hidden Markov models and Bayesian approaches to maximum likelihood estimation. Stat. Comput. 8, 145-158 (1998) · doi:10.1023/A:1008938201645
[33] Robert, C.P., Rydén, T., Titterington, D.M.: Bayesian inference in hidden Markov models through the reversible jump Markov chain Monte Carlo method. J. R. Stat. Soc. B 62(1), 57-65 (2000) · Zbl 0941.62090 · doi:10.1111/1467-9868.00219
[34] Rydén, T., Terasvirta, T., Asbrink, S.: Stylized facts of daily return series and the hidden Markov model. J. Appl. Econom. 13, 217-244 (1998) · doi:10.1002/(SICI)1099-1255(199805/06)13:3<217::AID-JAE476>3.0.CO;2-V
[35] Sansom, J., Thomson, P.: Fitting hidden semi-Markov models to breakpoint rainfall data. J. Appl. Probab. 38A, 142-157 (2001) · Zbl 1008.62110 · doi:10.1239/jap/1085496598
[36] Schmidler, S.C., Liu, J.S., Brutlag, D.L.: Bayesian segmentation of protein secondary structure. J. Comput. Biol. 7(1-2), 233-248 (2000) · doi:10.1089/10665270050081496
[37] Scott, S.: Bayesian methods for hidden Markov models: recursive computing in the 21st century. J. Am. Stat. Assoc. 97, 337-351 (2002) · Zbl 1073.65503 · doi:10.1198/016214502753479464
[38] Scott, S., Smyth, P.: The Markov modulated Poisson process and Markov Poisson cascade with applications to web traffic modelling. Bayesian Stat. 7, 671-680 (2003)
[39] Spiegelhalter, D.J., Best, N.G., Carlin, B.P., Van Der Linde, A.: Bayesian measures of model complexity and fit. J. R. Stat. Soc., Ser. B, Stat. Methodol. 64(4), 583-639 (2002) · Zbl 1067.62010 · doi:10.1111/1467-9868.00353
[40] Stephens, M.: Dealing with label switching in mixture models. J. R. Stat. Soc. B 62(4), 795-809 (2000) · Zbl 0957.62020 · doi:10.1111/1467-9868.00265
[41] Tokdar, S., Xi, P., Kelly, R., Kass, R.: Detection of bursts in extracellular spike trains using hidden semi-Markov point process models. J. Comput. Neurosci. 29, 203-212 (2010) · Zbl 1446.92182 · doi:10.1007/s10827-009-0182-2
[42] Yau, C., Papaspiliopoulos, O., Roberts, G.O., Holmes, C.: Bayesian non-parametric hidden Markov models with applications in genomics. J. R. Stat. Soc. B 73(1), 37-57 (2011) · Zbl 1411.62247 · doi:10.1111/j.1467-9868.2010.00756.x
[43] Yu, S.-Z.: Hidden semi-Markov models. Artif. Intell. 174, 215-243 (2010) · Zbl 1344.68181 · doi:10.1016/j.artint.2009.11.011
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.