Statistics and Computing

, Volume 16, Issue 2, pp 193–202 | Cite as

Markov Chain Monte Carlo in small worlds

  • Yongtao Guan
  • Roland Fleißner
  • Paul Joyce
  • Stephen M. Krone


As the number of applications for Markov Chain Monte Carlo (MCMC) grows, the power of these methods as well as their shortcomings become more apparent. While MCMC yields an almost automatic way to sample a space according to some distribution, its implementations often fall short of this task as they may lead to chains which converge too slowly or get trapped within one mode of a multi-modal space. Moreover, it may be difficult to determine if a chain is only sampling a certain area of the space or if it has indeed reached stationarity.

In this paper, we show how a simple modification of the proposal mechanism results in faster convergence of the chain and helps to circumvent the problems described above. This mechanism, which is based on an idea from the field of “small-world” networks, amounts to adding occasional “wild” proposals to any local proposal scheme. We demonstrate through both theory and extensive simulations, that these new proposal distributions can greatly outperform the traditional local proposals when it comes to exploring complex heterogenous spaces and multi-modal distributions. Our method can easily be applied to most, if not all, problems involving MCMC and unlike many other remedies which improve the performance of MCMC it preserves the simplicity of the underlying algorithm.


Markov Chain Monte Carlo Metropolis-Hastings algorithm Proposal distributions Small-world networks Importance sampling 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Chib S. and Greenberg E. 1995, Understanding the M etropolis-Hastings algorithm. The A merican statistician 49 : 327–335.CrossRefGoogle Scholar
  2. Chib S., Greenberg E., and Winkelmann R. 1998, Posterior simulation and B ayes factors in panel count data models. Journal of Econometrics 86 : 33–54.CrossRefGoogle Scholar
  3. Gamerman D. 1997, M arkov C hain M onte C arlo. Chapman & Hall, London.Google Scholar
  4. Geyer C. J. 1991, M arkov chain M onte C arlo maximum likelihood. In: E. M. Keramides (Ed.): Computing Science and Statistics: Proceedings of the 23rd Symposium on the Interface . Interface Foundation, Fairfax Station, pp. 156–163.Google Scholar
  5. Geyer C. J. 1992, Practical M arkov chain M onte C arlo. Statist. Sci. 7 : 473–483.Google Scholar
  6. Gilks W. R., Richardson S., and Spiegelhalter D. J. 1996, M arkov C hain M onte C arlo in practice. Chapman & Hall, London, 1st edition.Google Scholar
  7. Hastings W. K. 1970, Monte Carlo sampling methods using Markov chains and their applications. Biometrika 57 : 97–109.CrossRefzbMATHGoogle Scholar
  8. Jarner S. F. and Roberts G. O. 2001, Convergence of heavy tailed MCMC algorithms (preprint).Google Scholar
  9. Larget B. and Simon D. 1999, Markov chain Monte Carlo algorithms for the Bayesian analysis of phylogenetic trees. Mol. Biol. Evol. 16 : 750–759.Google Scholar
  10. Meyn S. P. and Tweedie R. L. 1996, Markov Chains and Stochastic Stability . Springer, New York.Google Scholar
  11. Press W. H., Flannery B. P., Teukolsky S. A., and Vetterling W. T. 1992, Numerical Recipes in C : The Art of Scientific Computing. Cambridge University Press, Cambridge.Google Scholar
  12. Rice J. A. 1994, Mathematical statistics and data analysis . Duxbury Press, Pacific Grove, 2nd edition.Google Scholar
  13. Roberts G. O., Gelman A., and Gilks W. R. 1997, Weak convergence and optimal scaling of random walk M etropolis algorithms. The Annals of Applied Probability 7 : 110–120.CrossRefMathSciNetGoogle Scholar
  14. Tierney L. 1994, M arkov chains for exploring posterior distributions. The Annals of Statistics 22 : 1701–1762.zbMATHMathSciNetGoogle Scholar
  15. Watts D. J. and Strogatz S. H. 1998, Collective dynamics of ‘small-world’ networks. Nature 393 : 440–442.CrossRefGoogle Scholar

Copyright information

© Springer Science + Business Media, LLC 2006

Authors and Affiliations

  • Yongtao Guan
    • 1
  • Roland Fleißner
    • 1
  • Paul Joyce
    • 1
    • 2
  • Stephen M. Krone
    • 1
  1. 1.Department of MathematicsUniversity of IdahoMoscow
  2. 2.Department of StatisticsUniversity of IdahoMoscow

Personalised recommendations