# zbMATH — the first resource for mathematics

Smoothing with couplings of conditional particle filters. (English) Zbl 1445.62245
Summary: In state-space models, smoothing refers to the task of estimating a latent stochastic process given noisy measurements related to the process. We propose an unbiased estimator of smoothing expectations. The lack-of-bias property has methodological benefits: independent estimators can be generated in parallel, and CI can be constructed from the central limit theorem to quantify the approximation error. To design unbiased estimators, we combine a generic debiasing technique for Markov chains, with a Markov chain Monte Carlo algorithm for smoothing. The resulting procedure is widely applicable and we show in numerical experiments that the removal of the bias comes at a manageable increase in variance. We establish the validity of the proposed estimators under mild assumptions. Numerical experiments are provided on toy models, including a setting of highly informative observations, and for a realistic Lotka-Volterra model with an intractable transition density.

##### MSC:
 62M20 Inference from stochastic processes and prediction 65C05 Monte Carlo methods 65C35 Stochastic particle methods
CODA; odeint
Full Text:
##### References:
 [1] Ahnert, K.; Mulansky, M., Odeint—Solving Ordinary Differential Equations in C++, AIP Conference Proceedings, 1389, 1586-1589 (2011) [2] Andrieu, C.; Doucet, A.; Holenstein, R., “Particle Markov Chain Monte Carlo” (with discussion), Journal of the Royal Statistical Society, Series B, 72, 357-385 (2010) [3] Andrieu, C.; Lee, A.; Vihola, M., “Uniform Ergodicity of the Iterated Conditional SMC and Geometric Ergodicity of Particle Gibbs Samplers,”, Bernoulli, 24, 842-872 (2018) · Zbl 1436.60069 [4] Bretó, C.; He, D.; Ionides, E. L.; King, A. A., “Time Series Analysis via Mechanistic Models,”, The Annals of Applied Statistics, 3, 319-348 (2009) · Zbl 1160.62080 [5] Briers, M.; Doucet, A.; Maskell, S., “Smoothing Algorithms for State-Space Models,”, Annals of the Institute of Statistical Mathematics, 62, 61-89 (2010) · Zbl 1422.62297 [6] Cappé, O.; Moulines, E.; Rydén, T., Inference in Hidden Markov Models (2005), New York: Springer-Verlag, New York · Zbl 1080.62065 [7] Carter, C. K.; Kohn, R., “On Gibbs Sampling for State Space Models,”, Biometrika, 81, 541-553 (1994) · Zbl 0809.62087 [8] Chopin, N.; Singh, S. S., “On Particle Gibbs Sampling,”, Bernoulli, 21, 1855-1883 (2015) · Zbl 1333.60164 [9] Del Moral, P.; Doucet, A.; Singh, S., “Forward Smoothing Using Sequential Monte Carlo,”, arXiv preprint arXiv:1012.5390 (2010) [10] Del Moral, P.; Murray, L. M., “Sequential Monte Carlo With Highly Informative Observations,”, SIAM/ASA Journal on Uncertainty Quantification, 3, 969-997 (2015) · Zbl 1327.65004 [11] Deligiannidis, G.; Doucet, A.; Pitt, M. K., “The Correlated Pseudo-Marginal, Method,” Journal of the Royal Statistical Society, Series B, 80, 839-870 (2018) · Zbl 1407.62074 [12] Diaconis, P.; Freedman, D., “Iterated Random Functions,”, SIAM Review, 41, 45-76 (1999) · Zbl 0926.60056 [13] Douc, R.; Garivier, A.; Moulines, E.; Olsson, J., “Sequential Monte Carlo Smoothing for General State Space Hidden Markov Models,”, The Annals of Applied Probability, 21, 2109-2145 (2011) · Zbl 1237.60026 [14] Douc, R.; Moulines, E.; Stoffer, D., Nonlinear Time Series: Theory, Methods and Applications with R Examples (2014), New York: Chapman and Hall, New York · Zbl 1306.62026 [15] Doucet, A.; de Freitas, N.; Gordon, N., Sequential Monte Carlo Methods in Practice (2001), New York: Springer-Verlag, New York · Zbl 0967.00022 [16] Fearnhead, P.; Wyncoll, D.; Tawn, J., “A Sequential Smoothing Algorithm With Linear Computational Cost,”, Biometrika, 97, 447-464 (2010) · Zbl 1406.62093 [17] Fulop, A.; Li, J., “Efficient Learning via Simulation: A Marginalized Resample-Move Approach,”, Journal of Econometrics, 176, 146-161 (2013) · Zbl 1284.62516 [18] Gelman, A.; Brooks, S.; Jones, G.; Meng, X., Handbook of Markov Chain Monte Carlo: Methods and Applications (2010), Boca Raton, FL: CRC Press, Boca Raton, FL [19] Gerber, M.; Chopin, N., “Sequential Quasi Monte Carlo,”, Journal of the Royal Statistical Society, Series B, 77, 509-579 (2015) · Zbl 1414.62109 [20] Glynn, P. W.; Rhee, C.-H, “Exact Estimation for Markov Chain Equilibrium Expectations,”, Journal of Applied Probability, 51A, 377-389 (2014) · Zbl 1312.65003 [21] Glynn, P. W.; Whitt, W., “The Asymptotic Efficiency of Simulation Estimators,”, Operations Research, 40, 505-520 (1992) · Zbl 0760.62009 [22] Godsill, S. J.; Doucet, A.; West, M., “Monte Carlo Smoothing for Nonlinear Time Series,”, Journal of the American Statistical Association, 99, 156-168 (2004) · Zbl 1089.62517 [23] Gordon, N.; Salmond, J.; Smith, A., A Novel Approach to Non-Linear/Non-Gaussian Bayesian State Estimation,”, IEEE Proceedings on Radar and Signal Processing, 140, 107-113 (1993) [24] Guarniero, P.; Johansen, A. M.; Lee, A., “The Iterated Auxiliary Particle Filter,”, Journal of the American Statistical Association, 112, 1636-1647 (2017) [25] Heng, J.; Bishop, A. N.; Deligiannidis, G.; Doucet, A., “Controlled Sequential Monte Carlo,”, arXiv preprint arXiv:1708.08396 (2017) [26] Huggins, J. H.; Roy, D. M., “Sequential Monte Carlo as Approximate Sampling: Bounds, Adaptive Resampling via $$####$$-ESS, and an Application to Particle Gibbs,”, arXiv preprint arXiv:1503.00966 (2018) [27] Jacob, P. E., “Sequential Bayesian Inference for Implicit Hidden Markov Models and Current Limitations,”, ESAIM: Proceedings and Surveys, 51, 24-48 (2015) · Zbl 1348.60106 [28] Jacob, P. E.; Lindsten, F.; Schön, T. B., “Coupling of Particle Filters,”, arXiv preprint arXiv:1606.01156 (2016) [29] Jacob, P. E.; Murray, L. M.; Rubenthaler, S., “Path Storage in the Particle Filter,”, Statistics and Computing, 25, 487-496 (2015) · Zbl 1332.62359 [30] Jacob, P. E.; O’Leary, J.; Atchadé, Y. F., “Unbiased Markov Chain Monte Carlo With Couplings,”, arXiv preprint arXiv:1708.03625 (2017) [31] Jasra, A.; Kamatani, K.; Law, K. J. H.; Zhou, Y., “Multilevel Particle Filters,”, SIAM Journal on Numerical Analysis, 55, 3068-3096 (2017) · Zbl 06819709 [32] Johansen, A. M.; Doucet, A., “A Note on Auxiliary Particle Filters,”, Statistics & Probability Letters, 78, 1498-1504 (2008) · Zbl 1152.62066 [33] Jones, E.; Parslow, J.; Murray, L., A Bayesian Approach to State and Parameter Estimation in a Phytoplankton-Zooplankton Model, Australian Meteorological and Oceanographic Journal, 59, 7-16 (2010) [34] Kantas, N.; Doucet, A.; Singh, S. S.; Maciejowski, J.; Chopin, N., “On Particle Methods for Parameter Estimation in State-Space Models,”, Statistical Science, 30, 328-351 (2015) · Zbl 1332.62096 [35] Kitagawa, G.; Sato, S., Sequential Monte Carlo Methods in Practice, “Monte Carlo Smoothing and Self-Organising State-Space Model, 177-195 (2001), Berlin: Springer, Berlin · Zbl 1056.93581 [36] Koskela, J.; Jenkins, P. A.; Johansen, A. M.; Spano, D., “Asymptotic Genealogies of Interacting Particle Systems With an Application to Sequential Monte Carlo,”, arXiv preprint arXiv:1804.01811 (2018) [37] Kuhlenschmidt, B.; Singh, S. S., “Stability of Conditional Sequential Monte Carlo,”, arXiv preprint arXiv:1806.06520 (2018) [38] Lee, A., “Towards Smooth Particle Filters for Likelihood Estimation With Multivariate Latent Variables,” (2008), University of British Columbia [39] Lee, A.; Doucet, A.; Łatuszyński, K., “Perfect Simulation Using Atomic Regeneration With Application to Sequential Monte Carlo,”, arXiv preprints arXiv:1407.5770 (2014) [40] Lee, A.; Singh, S. S.; Vihola, M., “Coupled Conditional Backward Sampling Particle Filter,”, arXiv preprint arXiv:1806.05852 (2018) [41] Lee, A.; Whiteley, N., “Forest Resampling for Distributed Sequential Monte Carlo,”, Statistical Analysis and Data Mining: The ASA Data Science Journal, 9, 230-248 (2015) [42] Lee, A.; Whiteley, N., “Variance Estimation and Allocation in the Particle Filter,”, arXiv preprint arXiv:1509.00394 (2015) [43] Lindsten, F.; Douc, R.; Moulines, E., “Uniform Ergodicity of the Particle Gibbs Sampler,”, Scandinavian Journal of Statistics, 42, 775-797 (2015) · Zbl 1360.60143 [44] Lindsten, F.; Jordan, M. I.; Schön, T. B., “Particle Gibbs With Ancestor Sampling,”, Journal of Machine Learning Research (JMLR), 15, 2145-2184 (2014) · Zbl 1319.60151 [45] Lindsten, F.; Schön, T. B., “Backward Simulation Methods for Monte Carlo Statistical Inference,”, Foundations and Trends in Machine Learning, 6, 1-143 (2013) · Zbl 1278.68016 [46] Lindvall, T., Lectures on the Coupling Method (2002), Chelmsford, MA: Courier Corporation, Chelmsford, MA · Zbl 1013.60001 [47] McLeish, D., “A General Method for Debiasing a Monte Carlo Estimator,”, Monte Carlo Methods and Applications, 17, 301-315 (2011) · Zbl 1238.65004 [48] Murray, L. M.; Lee, A.; Jacob, P. E., “Parallel Resampling in the Particle Filter,”, Journal of Computational and Graphical Statistics, 25, 789-805 (2016) [49] Murray, L. M.; Singh, S.; Jacob, P. E.; Lee, A., “Anytime Monte Carlo,”, arXiv preprint arXiv:1612.03319 (2016) [50] Olsson, J.; Cappé, O.; Douc, R.; Moulines, E., “Sequential Monte Carlo Smoothing With Application to Parameter Estimation in Nonlinear State Space Models,”, Bernoulli, 14, 155-179 (2008) · Zbl 1155.62055 [51] Olsson, J.; Douc, R., “Numerically Stable Online Estimation of Variance in Particle Filters,”, arXiv preprint arXiv:1701.01001 (2017) [52] Olsson, J.; Westerborn, J., “Efficient Particle-Based Online Smoothing in General Hidden Markov Models: The PaRIS Algorithm,”, Bernoulli, 23, 1951-1996 (2017) · Zbl 1392.62252 [53] Paige, B.; Wood, F.; Doucet, A.; Teh, Y. W., Advances in Neural Information Processing Systems, “Asynchronous Anytime Sequential Monte Carlo, 3410-3418 (2014) [54] Pitt, M. K. (2002) [55] Pitt, M. K.; Shephard, N., “Filtering via Simulation: Auxiliary Particle Filters,”, Journal of the American Statistical Association, 94, 590-599 (1999) · Zbl 1072.62639 [56] Plummer, M.; Best, N.; Cowles, K.; Vines, K., “CODA: Convergence Diagnosis and Output Analysis for MCMC,”, R News, 6, 7-11 (2006) [57] Rhee, C., “Unbiased Estimation with Biased Samplers,” (2013) [58] Rhee, C.; Glynn, P. W., A New Approach to Unbiased Estimation for SDE’s, Proceedings of the Winter Simulation Conference, 17-1-17-7 (2012) [59] Rosenthal, J. S., “Parallel Computing and Monte Carlo Algorithms,”, Far East Journal of Theoretical Statistics, 4, 207-236 (2000) · Zbl 1008.68160 [60] Ruiz, H.-C.; Kappen, H. J., “Particle Smoothing for Hidden Diffusion Processes: Adaptive Path Integral Smoother,”, IEEE Transactions on Signal Processing, 65, 3191-3203 (2017) · Zbl 1414.94525 [61] Sen, D.; Thiery, A. H.; Jasra, A., “On Coupling Particle Filter Trajectories,”, Statistics and Computing, 28, 461-475 (2018) · Zbl 1384.65006 [62] Singh, S. S.; Lindsten, F.; Moulines, E., “Blocking Strategies and Stability of Particle Gibbs Samplers,”, Biometrika, 104, 953-969 (2017) · Zbl 07072338 [63] Taghavi, E.; Lindsten, F.; Svensson, L.; Schön, T. B., “Adaptive Stopping for Fast Particle Smoothing,”, Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 6293-6297 (2013), Vancouver, Canada [64] Vats, D.; Flegal, J. M.; Jones, G. L., “Strong Consistency of Multivariate Spectral Variance Estimators in Markov Chain Monte Carlo,”, Bernoulli, 24, 1860-1909 (2018) · Zbl 1419.62141 [65] Vihola, M., “Unbiased Estimators and Multilevel Monte Carlo,”, Operations Research, arXiv preprint arXiv, 1512, 01022 (2017) [66] Whiteley, N., “Comment on Particle Markov Chain Monte Carlo by Andrieu, Doucet and Holenstein,”, Journal of the Royal Statistical Society, Series B, 72, 357-385 (2010) [67] Whiteley, N., “Stability Properties of Some Particle Filters,”, The Annals of Applied Probability, 23, 2500-2537 (2013) · Zbl 1296.60098 [68] Whiteley, N.; Lee, A.; Heine, K., “On the Role of Interaction in Sequential Monte Carlo Algorithms,”, Bernoulli, 22, 494-529 (2016) · Zbl 1388.65009
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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.