On the importance function in splitting simulation

The splitting method is a simulation technique for the estimation of very small probabilities. In this technique, the sample paths are split into multiple copies, at various stages in the simulation. Of vital importance to the efficiency of the method is the Importance Function (IF). This function governs the placement of the thresholds or surfaces at which the paths are split We derive a characterisation of the optimal IF and show that for multi-dimensional models the natural choice for the IF is usually not optimal. We also show how nearly optimal splitting surfaces can be derived or simulated using reverse time analysis. Our numerical experiments illustrate that by using the optimal IF, one can obtain a significant improvement in simulation efficiency.

[1]  Marnix J. J. Garvels,et al.  The splitting method in rare event simulation , 2000 .

[2]  Paul Glasserman,et al.  Splitting for rare event simulation: analysis of simple cases , 1996, Winter Simulation Conference.

[3]  Dirk P. Kroese,et al.  Efficient simulation of a tandem Jackson network , 1999, WSC.

[4]  José Villén-Altamirano,et al.  RESTART: a straightforward method for fast simulation of rare events , 1994, Proceedings of Winter Simulation Conference.

[5]  P. Glasserman,et al.  A Look At Multilevel Splitting , 1998 .

[6]  J. Townsend,et al.  Efficient rare event simulation using DPR for multidimensional parameter spaces , 1998 .

[7]  Paul Glasserman,et al.  Multilevel Splitting for Estimating Rare Event Probabilities , 1999, Oper. Res..

[8]  B. Ripley,et al.  Robust Statistics , 2018, Wiley Series in Probability and Statistics.

[9]  Dirk P. Kroese,et al.  A comparison of RESTART implementations , 1998, 1998 Winter Simulation Conference. Proceedings (Cat. No.98CH36274).

[10]  Frank Kelly,et al.  Reversibility and Stochastic Networks , 1979 .

[11]  Michael R. Frater,et al.  Optimally efficient estimation of the statistics of rare events in queueing networks , 1991 .

[12]  Paul Glasserman,et al.  Analysis of an importance sampling estimator for tandem queues , 1995, TOMC.

[13]  J. Keith Townsend,et al.  Rare event simulation of delay in packet switching networks using DPR-based splitting , 1999, WSC '99.

[14]  Manuel VILLÉN-ALTAMIRANO Telefónica,et al.  On The Efficiency of RESTART , .

[15]  Ward Whitt,et al.  The Asymptotic Efficiency of Simulation Estimators , 1992, Oper. Res..

[16]  C. Gorg,et al.  COMPARISON AND OPTIMIZATION OF RESTART RUN TIME STRATEGIES , 1998 .

[17]  P. Glasserman,et al.  A large deviations perspective on the efficiency of multilevel splitting , 1998, IEEE Trans. Autom. Control..

[18]  Dirk P. Kroese,et al.  On the entrance distribution in RESTART simulation , 1999 .

[19]  Jean Walrand,et al.  A quick simulation method for excessive backlogs in networks of queues , 1989 .

[20]  P. Tsoucas Rare events in series of queues , 1992 .

[21]  Manuel Villén-Altamirano,et al.  Enhancement of the Accelerated Simulation Method RESTART by Considering Multiple Thresholds , 1994 .