Hostname: page-component-cd9895bd7-7cvxr Total loading time: 0 Render date: 2024-12-27T14:02:57.064Z Has data issue: false hasContentIssue false
  • English
  • Français

Efficient Simulation of Large Deviation Events for Sums of Random Vectors Using Saddle-Point Representations

Part of: Distribution theory - Probability Limit theorems Numerical methods in Fourier analysis Probabilistic methods, simulation and stochastic differential equations

Published online by Cambridge University Press:  30 January 2018

Ankush Agarwal ,
Santanu Dey  and
Sandeep Juneja
Ankush Agarwal*
Affiliation:
Tata Institute of Fundamental Research
Santanu Dey*
Affiliation:
Tata Institute of Fundamental Research
Sandeep Juneja*
Affiliation:
Tata Institute of Fundamental Research
*
∗ Postal address: STCS, Tata Institute of Fundamental Research, 1 Homi Bhabha Road, Mumbai, India.
∗ Postal address: STCS, Tata Institute of Fundamental Research, 1 Homi Bhabha Road, Mumbai, India.
∗ Postal address: STCS, Tata Institute of Fundamental Research, 1 Homi Bhabha Road, Mumbai, India.
Rights & Permissions [Opens in a new window]

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

We consider the problem of efficient simulation estimation of the density function at the tails, and the probability of large deviations for a sum of independent, identically distributed (i.i.d.), light-tailed, and nonlattice random vectors. The latter problem besides being of independent interest, also forms a building block for more complex rare event problems that arise, for instance, in queueing and financial credit risk modeling. It has been extensively studied in the literature where state-independent, exponential-twisting-based importance sampling has been shown to be asymptotically efficient and a more nuanced state-dependent exponential twisting has been shown to have a stronger bounded relative error property. We exploit the saddle-point-based representations that exist for these rare quantities, which rely on inverting the characteristic functions of the underlying random vectors. These representations reduce the rare event estimation problem to evaluating certain integrals, which may via importance sampling be represented as expectations. Furthermore, it is easy to identify and approximate the zero-variance importance sampling distribution to estimate these integrals. We identify such importance sampling measures and show that they possess the asymptotically vanishing relative error property that is stronger than the bounded relative error property. To illustrate the broader applicability of the proposed methodology, we extend it to develop an asymptotically vanishing relative error estimator for the practically important expected overshoot of sums of i.i.d. random variables.

Keywords

Rare event simulationimportance samplingsaddle-point approximationFourier inversionlarge deviations

MSC classification

Primary: 65C05: Monte Carlo methods 60E10: Characteristic functions; other transforms 60F10: Large deviations
Secondary: 65C50: Other computational problems in probability 65T99: None of the above, but in this section
Type
Research Article
Information
Journal of Applied Probability , Volume 50 , Issue 3 , September 2013 , pp. 703 - 720
Copyright
© Applied Probability Trust 

References

Asmussen, S. and Glynn, P. W. (2008). Stochastic Simulation: Algorithms and Analysis. Springer, New York.Google Scholar
Blanchet, J. H., Leder, K. and Glynn, P. W. (2009). Strongly efficient algorithms for light-tailed random walks: An old folk song sung to a faster new tune. In Monte Carlo and Quasi-Monte Carlo Methods 2008, Springer, Berlin, pp. 227–248.Google Scholar
Bucklew, J. A. (2004). Introduction to Rare Event Simulation. Springer, New York.Google Scholar
Bucklew, J. A., Ney, P. and Sadowsky, J. S. (1990). Monte Carlo simulation and large deviations theory for uniformly recurrent Markov chains. J. Appl. Prob. 27, 44–59.Google Scholar
Butler, R. W. (2007). Saddlepoint Approximations with Applications. Cambridge University Press.Google Scholar
Daniels, H. E. (1954). Saddlepoint approximation in statistics. Ann. Math. Statist. 25, 631–650.Google Scholar
Dey, S. and Juneja, S. (2011). Efficient estimation of density and probability of large deviations of sum of IID random variables. In Proc. 2011 Winter Simulation Conf., IEEE, pp. 3800–3811.Google Scholar
Dey, S., Juneja, S. and Agarwal, A. (2012). Efficient simulation of density and probability of large deviations of sum of random vectors using saddle point representations. Preprint. Available at http://arxiv.org/abs/1203.0817v2.Google Scholar
Dieker, A. B. and Mandjes, M. (2005). On asymptotically efficient simulation of large deviation probabilities. Adv. Appl. Prob. 37, 539–552.CrossRefGoogle Scholar
Feller, W. (1971). An Introduction to Probability Theory and Its Applications, Vol. II. John Wiley, New York.Google Scholar
Glasserman, P. and Juneja, S. (2008). Uniformly efficient importance sampling for the tail distribution of sums of random variables. Math. Operat. Res. 33, 36–50.Google Scholar
Glasserman, P. and Li, J. (2005). Importance sampling for portfolio credit risk. Manag. Sci. 51, 1643–1656.CrossRefGoogle Scholar
Glasserman, P. and Wang, Y. (1997). Counterexamples in importance sampling for large deviations probabilities. Ann. Appl. Prob. 7, 731–746.Google Scholar
Jensen, J. L. (1995). Saddlepoint Approximations. Oxford University Press, New York.Google Scholar
Juneja, S. and Shahabuddin, P. (2006). Rare event simulation techniques. In Handbooks in Operations Research and Management Science, Vol. 13, eds Henderson, S. G. and Nelson, B. L., North-Holland, Amsterdam, pp. 291–350.Google Scholar
Lugannani, R. and Rice, S. (1980). Saddle point approximation for the distribution of the sum of independent random variables. Adv. Appl. Prob. 12, 475–490.Google Scholar
Parekh, S. and Walrand, J. (1989). A quick simulation method for excessive backlogs in networks of queues. IEEE Trans. Automatic Control 34, 54–66.Google Scholar
Sadowsky, J. S. (1996). On Monte Carlo estimation of large deviations probabilities. Ann. Appl. Prob. 6, 399–422.CrossRefGoogle Scholar
Sadowsky, J. S. and Bucklew, J. A. (1990). On large deviations theory and asymptotically efficient Monte Carlo simulation estimation. IEEE Trans. Inf. Theory 36, 579–588.Google Scholar