Statistics and Computing

, Volume 23, Issue 2, pp 271–285 | Cite as

Markov chain importance sampling with applications to rare event probability estimation

  • Zdravko I. Botev
  • Pierre L’Ecuyer
  • Bruno Tuffin
Article

Abstract

We present a versatile Monte Carlo method for estimating multidimensional integrals, with applications to rare-event probability estimation. The method fuses two distinct and popular Monte Carlo simulation methods—Markov chain Monte Carlo and importance sampling—into a single algorithm. We show that for some applied numerical examples the proposed Markov Chain importance sampling algorithm performs better than methods based solely on importance sampling or MCMC.

Keywords

Importance sampling Nonparametric Minimum variance density Rare-event probability Variance reduction 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Asmussen, S., Glynn, P.W.: Stochastic Simulation: Algorithms and Analysis. Springer, New York (2007) MATHGoogle Scholar
  2. Asmussen, S., Blanchet, J., Juneja, S., Rojas-Nandayapa, L.: Efficient simulation of tail probabilities of sums of correlated lognormals. Ann. Oper. Res. (2009). doi:10.1007/s10479-009-0658-5 MATHGoogle Scholar
  3. Bassamboo, A., Juneja, S., Zeevi, A.: Portfolio credit risk with extremal dependence: Asymptotic analysis and efficient simulation. Oper. Res. 56(3), 593–606 (2008) MathSciNetMATHCrossRefGoogle Scholar
  4. Besag, J.: Statistical analysis of non-lattice data. Statistician 24(3), 179–195 (1975) MathSciNetCrossRefGoogle Scholar
  5. Blanchet, J.H., Glynn, P.W.: Efficient rare-event simulation for the maximum of heavy-tailed random walks. Ann. Appl. Probab. 18, 1351–1378 (2008) MathSciNetMATHCrossRefGoogle Scholar
  6. Botev, Z.I., Kroese, D.P.: Efficient Monte Carlo simulation via the generalized splitting method. Stat. Comput. (2010). doi:10.1007/s11222-010-9201-4 Google Scholar
  7. Botev, Z.I., L’Ecuyer, P., Tuffin, B.: Importance sampling method based on a one-step look-ahead density from a Markov chain. In: Proceedings of the 2011 Winter Simulation Conference, Phoenix, AZ (2011) Google Scholar
  8. Botev, Z.I., L’Ecuyer, P., Rubino, G., Simard, R., Tuffin, B.: Static network reliability estimation via generalized splitting. INFORMS Journal on Computing (2012, to appear) Google Scholar
  9. Brereton, T.J., Chan, J.C.C., Kroese, D.P.: Fitting mixture importance sampling distributions via improved cross-entropy. In: Proceedings of the 2011 Winter Simulation Conference, Phoenix, AZ (2011) Google Scholar
  10. Bucklew, J.: Introduction to Rare-Event Simulation. Springer, New York (2004) MATHGoogle Scholar
  11. Cancela, H., El Khadiri, M., Rubino, G.: Rare event analysis by Monte Carlo techniques in static models. In: Rubino, G., Tuffin, B. (eds.) Rare Event Simulation Using Monte Carlo Methods, pp. 145–170. Wiley, New York (2009a). Chap. 7 CrossRefGoogle Scholar
  12. Cancela, H., L’Ecuyer, P., Lee, M., Rubino, G., Tuffin, B.: Analysis and improvements of path-based methods for Monte Carlo reliability evaluation of static models. In: Faulin, J., Juan, A.A., Martorell, S., Ramirez-Marquez, E. (eds.) Simulation Methods for Reliability and Availability of Complex Systems, pp. 65–84. Springer, Berlin (2009b) Google Scholar
  13. Chan, J.C.C., Kroese, D.P.: Improved cross-entropy method for estimation. Stat. Comput. (2011). doi:10.1007/s11222-011-9275-7 Google Scholar
  14. Chan, J.C.C., Glynn, P.W., Kroese, D.P.: A comparison of cross-entropy and variance minimization strategies. J. Appl. Probab. 48A, 183–194 (2011) MathSciNetMATHCrossRefGoogle Scholar
  15. Chib, S.: Marginal likelihood from the Gibbs output. J. Am. Stat. Assoc. 90(432), 1313–1321 (1995) MathSciNetMATHCrossRefGoogle Scholar
  16. Dupuis, P., Leder, K., Wang, H.: Importance sampling for sums of random variables with regularly varying tails. ACM Trans. Model. Comput. Simul. 17(3), 14 (2006) CrossRefGoogle Scholar
  17. Gertsbakh, I.B., Shpungin, Y.: Models of Network Reliability. CRC Press, Boca Raton (2010) MATHGoogle Scholar
  18. Glasserman, P.: Monte Carlo Methods in Financial Engineering. Springer, New York (2004) MATHGoogle Scholar
  19. Glasserman, P., Wang, Y.: Counterexamples in importance sampling for large deviations probabilities. Ann. Appl. Probab. 7(3), 731–746 (1997) MathSciNetMATHCrossRefGoogle Scholar
  20. Kroese, D.P., Taimre, T., Botev, Z.I.: Handbook of Monte Carlo Methods. Wiley, New York (2011) MATHCrossRefGoogle Scholar
  21. L’Ecuyer, P., Tuffin, B.: Approximate zero-variance simulation. In: Proceedings of the 2008 Winter Simulation Conference, pp. 170–181. IEEE Press, New York (2008) CrossRefGoogle Scholar
  22. L’Ecuyer, P., Blanchet, J.H., Tuffin, B., Glynn, P.W.: Asymptotic robustness of estimators in rare-event simulation. ACM Trans. Model. Comput. Simul. 20(1), 6 (2010) Google Scholar
  23. Liu, J.S.: Monte Carlo Strategies in Scientific Computing. Springer, New York (2001) MATHGoogle Scholar
  24. Metropolis, N., Rosenbluth, A.W., Rosenbluth, M.N., Teller, A.H., Teller, E.: Equations of state calculations by fast computing machines. J. Chem. Phys. 21(6), 1087–1092 (1953) CrossRefGoogle Scholar
  25. Philippe, A.: Simulation of right and left truncated gamma distribution by mixtures. Stat. Comput. 7, 173–181 (1997) CrossRefGoogle Scholar
  26. Robert, C.P., Casella, G.: Monte Carlo Statistical Methods, 2nd edn. Springer, New York (2004) MATHGoogle Scholar
  27. Rubino, G., Tuffin, B. (eds.): Rare Event Simulation Using Monte Carlo Methods. Wiley, New York (2009) MATHGoogle Scholar
  28. Rubinstein, R.Y., Kroese, D.P.: The Cross-Entropy Method: A Unified Approach to Combinatorial Optimization, Monte-Carlo Simulation and Machine Learning. Springer, New York (2004) MATHGoogle Scholar
  29. Rubinstein, R.Y., Kroese, D.P.: Simulation and the Monte Carlo Method, 2nd edn. Wiley, New York (2007) CrossRefGoogle Scholar
  30. Zhang, P.: Nonparametric importance sampling. J. Am. Stat. Assoc. 91(435), 1245–1253 (1996) MATHCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Zdravko I. Botev
    • 1
  • Pierre L’Ecuyer
    • 2
  • Bruno Tuffin
    • 3
  1. 1.School of Mathematics and StatisticsUniversity of New South WalesSydneyAustralia
  2. 2.Department of Computer Science and Operations ResearchUniversité de MontréalMontréalCanada
  3. 3.INRIA Rennes Bretagne-AtlantiqueRennes CedexFrance

Personalised recommendations