Negative association, ordering and convergence of resampling methods

We study convergence and convergence rates for resampling schemes. Our first main result is a general consistency theorem based on the notion of negative association, which is applied to establish the almost-sure weak convergence of measures output from Kitagawa's (1996) stratified resampling method. Carpenter et al's (1999) systematic resampling method is similar in structure but can fail to converge depending on the order of the input samples. We introduce a new resampling algorithm based on a stochastic rounding technique of Srinivasan (2001), which shares some attractive properties of systematic resampling, but which exhibits negative association and therefore converges irrespective of the order of the input samples. We confirm a conjecture made by Kitagawa (1996) that ordering input samples by their states in $\mathbb{R}$ yields a faster rate of convergence; we establish that when particles are ordered using the Hilbert curve in $\mathbb{R}^d$, the variance of the resampling error is ${\scriptscriptstyle\mathcal{O}}(N^{-(1+1/d)})$ under mild conditions, where $N$ is the number of particles. We use these results to establish asymptotic properties of particle algorithms based on resampling schemes that differ from multinomial resampling.

[1]  Zhijian He,et al.  Extensible grids: uniform sampling on a space filling curve , 2014, ArXiv.

[2]  Eric Moulines,et al.  Comparison of resampling schemes for particle filtering , 2005, ISPA 2005. Proceedings of the 4th International Symposium on Image and Signal Processing and Analysis, 2005..

[3]  P. Moral,et al.  On a Class of Genealogical and Interacting Metropolis Models , 2003 .

[4]  P. Bickel,et al.  Uniform Convergence of Probability Measures on Classes of Functions , 2008 .

[5]  M. Stein Large sample properties of simulations using latin hypercube sampling , 1987 .

[6]  H. Kunsch Recursive Monte Carlo filters: Algorithms and theoretical analysis , 2006, math/0602211.

[7]  M. Pitt,et al.  Filtering via Simulation: Auxiliary Particle Filters , 1999 .

[8]  Colas Schretter,et al.  Van der Corput and Golden Ratio Sequences Along the Hilbert Space-Filling Curve , 2014, MCQMC.

[9]  Lauwerens Kuipers,et al.  Uniform distribution of sequences , 1974 .

[10]  Stéphan Clémençon,et al.  Empirical Processes in Survey Sampling with (Conditional) Poisson Designs , 2017 .

[11]  Devdatt P. Dubhashi,et al.  Balls and bins: A study in negative dependence , 1996, Random Struct. Algorithms.

[12]  Approximation of continuous functions on Fréchet spaces , 2010, 1005.1814.

[13]  R. Douc,et al.  Optimality of the auxiliary particle filter , 2009 .

[14]  Jun S. Liu,et al.  Sequential Monte Carlo methods for dynamic systems , 1997 .

[15]  Eric Moulines,et al.  Adaptive methods for sequential importance sampling with application to state space models , 2008, 2008 16th European Signal Processing Conference.

[16]  P. L’Ecuyer,et al.  Variance Reduction via Lattice Rules , 1999 .

[17]  P. Fearnhead,et al.  Improved particle filter for nonlinear problems , 1999 .

[18]  Yves Tillé,et al.  Sampling Algorithms , 2011, International Encyclopedia of Statistical Science.

[19]  J. Deville,et al.  Unequal probability sampling without replacement through a splitting method , 1998 .

[20]  Hrvoje Babic,et al.  Image and Signal Processing and Analysis , 2001 .

[21]  A. Doucet,et al.  A note on auxiliary particle filters , 2008 .

[22]  Qi-Man Shao,et al.  A Comparison Theorem on Moment Inequalities Between Negatively Associated and Independent Random Variables , 2000 .

[23]  Aravind Srinivasan,et al.  Distributions on level-sets with applications to approximation algorithms , 2001, Proceedings 2001 IEEE International Conference on Cluster Computing.

[24]  A. V. D. Vaart,et al.  Asymptotic Statistics: Frontmatter , 1998 .

[25]  K. Joag-dev,et al.  Negative Association of Random Variables with Applications , 1983 .

[26]  N. Chopin,et al.  Sequential Quasi-Monte Carlo , 2014, 1402.4039.

[27]  Gerhard Zumbusch,et al.  Parallel Multilevel Methods , 2003 .

[28]  On Positive In uence and Negative Dependence , 2018 .

[29]  N. Chopin,et al.  Convergence of sequential quasi-Monte Carlo smoothing algorithms , 2015, 1506.06117.

[30]  A. Doucet,et al.  Particle Markov chain Monte Carlo methods , 2010 .

[31]  Jonathan Cutler,et al.  Negative Dependence and Srinivasan's Sampling Process , 2011, Comb. Probab. Comput..

[32]  N. Chopin Central limit theorem for sequential Monte Carlo methods and its application to Bayesian inference , 2004, math/0508594.

[33]  Harald Niederreiter,et al.  Random number generation and Quasi-Monte Carlo methods , 1992, CBMS-NSF regional conference series in applied mathematics.

[34]  Haikady N. Nagaraja,et al.  Inference in Hidden Markov Models , 2006, Technometrics.

[35]  N. Gordon,et al.  Novel approach to nonlinear/non-Gaussian Bayesian state estimation , 1993 .

[36]  Desh Ranjan,et al.  Positive Influence and Negative Dependence , 2006, Combinatorics, Probability and Computing.

[37]  Dan Crisan,et al.  Particle Filters - A Theoretical Perspective , 2001, Sequential Monte Carlo Methods in Practice.

[38]  W. G. Madow On the Theory of Systematic Sampling, III. Comparison of Centered and Random Start Systematic Sampling , 1953 .

[39]  S. Jaffard,et al.  Pointwise smoothness of space-filling functions , 2009 .

[40]  Arnaud Doucet,et al.  A survey of convergence results on particle filtering methods for practitioners , 2002, IEEE Trans. Signal Process..

[41]  C. Aistleitner,et al.  Functions of bounded variation, signed measures, and a general Koksma-Hlawka inequality , 2014, 1406.0230.

[42]  Tian Wang,et al.  Improved Particle Filter , 2014, CIT 2014.

[43]  Steven A. Orszag,et al.  CBMS-NSF REGIONAL CONFERENCE SERIES IN APPLIED MATHEMATICS , 1978 .

[44]  G. Pólya,et al.  Über den zentralen Grenzwertsatz der Wahrscheinlichkeitsrechnung und das Momentenproblem , 1920 .