Bernoulli

  • Bernoulli
  • Volume 25, Number 4B (2019), 3832-3863.

Consistent estimation of the spectrum of trace class Data Augmentation algorithms

Saptarshi Chakraborty and Kshitij Khare

Full-text: Access denied (no subscription detected)

We're sorry, but we are unable to provide you with the full text of this article because we are not able to identify you as a subscriber. If you have a personal subscription to this journal, then please login. If you are already logged in, then you may need to update your profile to register your subscription. Read more about accessing full-text

Abstract

Markov chain Monte Carlo is widely used in a variety of scientific applications to generate approximate samples from intractable distributions. A thorough understanding of the convergence and mixing properties of these Markov chains can be obtained by studying the spectrum of the associated Markov operator. While several methods to bound/estimate the second largest eigenvalue are available in the literature, very few general techniques for consistent estimation of the entire spectrum have been proposed. Existing methods for this purpose require the Markov transition density to be available in closed form, which is often not true in practice, especially in modern statistical applications. In this paper, we propose a novel method to consistently estimate the entire spectrum of a general class of Markov chains arising from a popular and widely used statistical approach known as Data Augmentation. The transition densities of these Markov chains can often only be expressed as intractable integrals. We illustrate the applicability of our method using real and simulated data.

Article information

Source
Bernoulli, Volume 25, Number 4B (2019), 3832-3863.

Dates
Received: November 2017
Revised: July 2018
First available in Project Euclid: 25 September 2019

Permanent link to this document
https://projecteuclid.org/euclid.bj/1569398786

Digital Object Identifier
doi:10.3150/19-BEJ1112

Mathematical Reviews number (MathSciNet)
MR4010974

Zentralblatt MATH identifier
07110157

Keywords
Data Augmentation algorithms eigenvalues of Markov operators MCMC convergence trace class Markov operators

Citation

Chakraborty, Saptarshi; Khare, Kshitij. Consistent estimation of the spectrum of trace class Data Augmentation algorithms. Bernoulli 25 (2019), no. 4B, 3832--3863. doi:10.3150/19-BEJ1112. https://projecteuclid.org/euclid.bj/1569398786


Export citation

References

  • Adamczak, R. and Bednorz, W. (2015). Some remarks on MCMC estimation of spectra of integral operators. Bernoulli 21 2073–2092.
  • Albert, J.H. and Chib, S. (1993). Bayesian analysis of binary and polychotomous response data. J. Amer. Statist. Assoc. 88 669–679.
  • Asmussen, S. and Glynn, P.W. (2011). A new proof of convergence of MCMC via the ergodic theorem. Statist. Probab. Lett. 81 1482–1485.
  • Athreya, K.B. and Atuncar, G.S. (1998). Kernel estimation for real-valued Markov chains. Sankhya, Ser. A 60 1–17.
  • Bennett, C.H. (1976). Efficient estimation of free energy differences from Monte Carlo data. J. Comput. Phys. 22 245–268.
  • Canty, A. and Ripley, B.D. (2017). boot: Bootstrap R (S-Plus) Functions. R package version 1.3-19.
  • Chakraborty, S. and Khare, K. (2017). Convergence properties of Gibbs samplers for Bayesian probit regression with proper priors. Electron. J. Stat. 11 177–210.
  • Chakraborty, S. and Khare, K. (2019). Supplement to “Consistent estimation of the spectrum of trace class data augmentation algorithms.” DOI:10.3150/19-BEJ1112SUPP.
  • Choi, H.M. and Hobert, J.P. (2013). The Polya-gamma Gibbs sampler for Bayesian logistic regression is uniformly ergodic. Electron. J. Stat. 7 2054–2064.
  • Choi, H.M. and Román, J.C. (2017). Analysis of Polya-Gamma Gibbs sampler for Bayesian logistic analysis of variance. Electron. J. Stat. 11 326–337.
  • Conway, J.B. (1990). A Course in Functional Analysis, 2nd ed. Graduate Texts in Mathematics 96. New York: Springer.
  • Davison, A.C. and Hinkley, D.V. (1997). Bootstrap Methods and Their Application. Cambridge Series in Statistical and Probabilistic Mathematics 1. Cambridge: Cambridge Univ. Press.
  • Diaconis, P., Khare, K. and Saloff-Coste, L. (2008). Gibbs sampling, exponential families and orthogonal polynomials. Statist. Sci. 23 151–178.
  • Diaconis, P. and Saloff-Coste, L. (1993). Comparison techniques for random walk on finite groups. Ann. Probab. 21 2131–2156.
  • Diaconis, P. and Saloff-Coste, L. (1996). Nash inequalities for finite Markov chains. J. Theoret. Probab. 9 459–510.
  • Diaconis, P. and Stroock, D. (1991). Geometric bounds for eigenvalues of Markov chains. Ann. Appl. Probab. 1 36–61.
  • Eddelbuettel, D. (2013). Seamless R and C$+$$+$ Integration with Rcpp. New York: Springer.
  • François, O. (2000). Geometric inequalities for the eigenvalues of concentrated Markov chains. J. Appl. Probab. 37 15–28.
  • Frühwirth-Schnatter, S. (2001). Markov chain Monte Carlo estimation of classical and dynamic switching and mixture models. J. Amer. Statist. Assoc. 96 194–209.
  • Garren, S.T. and Smith, R.L. (2000). Estimating the second largest eigenvalue of a Markov transition matrix. Bernoulli 6 215–242.
  • Hobert, J. P., Jung, Y. J., Khare, K. and Qin, Q. (2015). Convergence analysis of the Data Augmentation algorithm for Bayesian linear regression with non-Gaussian errors. ArXiv e-prints.
  • Hobert, J.P., Roy, V. and Robert, C.P. (2011). Improving the convergence properties of the data augmentation algorithm with an application to Bayesian mixture modeling. Statist. Sci. 26 332–351.
  • Hoffman, A.J. and Wielandt, H.W. (1953). The variation of the spectrum of a normal matrix. Duke Math. J. 20 37–39.
  • Jones, G.L. and Hobert, J.P. (2001). Honest exploration of intractable probability distributions via Markov chain Monte Carlo. Statist. Sci. 16 312–334.
  • Jörgens, K. (1982). Linear Integral Operators. Surveys and Reference Works in Mathematics 7. Boston, MA–London: Pitman.
  • Khare, K. and Hobert, J.P. (2011). A spectral analytic comparison of trace-class data augmentation algorithms and their sandwich variants. Ann. Statist. 39 2585–2606.
  • Khare, K. and Zhou, H. (2009). Rates of convergence of some multivariate Markov chains with polynomial eigenfunctions. Ann. Appl. Probab. 19 737–777.
  • Koltchinskii, V. and Giné, E. (2000). Random matrix approximation of spectra of integral operators. Bernoulli 6 113–167.
  • Lawler, G.F. and Sokal, A.D. (1988). Bounds on the $L^{2}$ spectrum for Markov chains and Markov processes: A generalization of Cheeger’s inequality. Trans. Amer. Math. Soc. 309 557–580.
  • Liu, J.S., Wong, W.H. and Kong, A. (1994). Covariance structure of the Gibbs sampler with applications to the comparisons of estimators and augmentation schemes. Biometrika 81 27–40.
  • Meng, X.-L. and Wong, W.H. (1996). Simulating ratios of normalizing constants via a simple identity: A theoretical exploration. Statist. Sinica 6 831–860.
  • Pal, S., Khare, K. and Hobert, J.P. (2017). Trace class Markov chains for Bayesian inference with generalized double Pareto shrinkage priors. Scand. J. Stat. 44 307–323.
  • Polson, N.G., Scott, J.G. and Windle, J. (2013). Bayesian inference for logistic models using Pólya-Gamma latent variables. J. Amer. Statist. Assoc. 108 1339–1349.
  • Qin, Q. and Hobert, J.P. (2018). Trace-class Monte Carlo Markov chains for Bayesian multivariate linear regression with non-Gaussian errors. J. Multivariate Anal. 166 335–345.
  • Qin, Q., Hobert, J.P. and Khare, K. (2017). Estimating the spectral gap of a trace-class Markov operator. Preprint. Available at arXiv:1704.00850.
  • R Core Team (2015). R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing.
  • Raftery, A.E. and Lewis, S. (1992). How many iterations in the Gibbs sampler. Bayesian Stat. 4 763–773.
  • Rajaratnam, B., Sparks, D., Khare, K. and Zhang, L. (2017). Scalable Bayesian shrinkage and uncertainty quantification in high-dimensional regression. ArXiv e-prints.
  • Retherford, J.R. (1993). Hilbert Space: Compact Operators and the Trace Theorem. London Mathematical Society Student Texts 27. Cambridge: Cambridge Univ. Press.
  • Rosenthal, J.S. (1995). Minorization conditions and convergence rates for Markov chain Monte Carlo. J. Amer. Statist. Assoc. 90 558–566.
  • Roy, V. (2012). Convergence rates for MCMC algorithms for a robust Bayesian binary regression model. Electron. J. Stat. 6 2463–2485.
  • Saloff-Coste, L. (2004). Total variation lower bounds for finite Markov chains: Wilson’s lemma. In Random Walks and Geometry 515–532. Berlin: de Gruyter.
  • Sinclair, A. and Jerrum, M. (1989). Approximate counting, uniform generation and rapidly mixing Markov chains. Inform. and Comput. 82 93–133.
  • Wickham, H. (2007). Reshaping data with the reshape package. J. Stat. Softw. 21 1–20.
  • Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. New York: Springer.
  • Yuen, W.K. (2000). Applications of geometric bounds to the convergence rate of Markov chains on $\mathbf{R}^{n}$. Stochastic Process. Appl. 87 1–23.

Supplemental materials

  • Supplement to “Consistent estimation of the spectrum of trace class data augmentation algorithms”. The supplement provides proofs of the theorems and lemmas introduced in this article.