Abstract

The RESTART method is a widely applicable simulation technique for the estimation of rare event probabilities. The method is based on the idea to restart the simulation at certain intermediate stages, in order to generate more occurrences of the rare event. In many cases we are interested in the (rare) event that a real-valued function of some Markov process exceeds a high level. We explore the possibility of speeding up the RESTART method for such models by using certain conditional distributions of the Markov process, called 'entrance distributions'. As a by-product, we show, for certain models, a remarkable relationship between the entrance distribution and the optimal exponential change of measure used in the Importance Sampling method, the other main rare event simulation technique. We find that using the entrance distribution in the RESTART method will usually yield a significant reduction in simulation effort in situations where the method can be applied. The standard RESTART method is more robust, but is not as efficient as the new method since it does not use any information on the system we are simulating.
Original languageUndefined
Place of PublicationEnschede
PublisherCentre for Telematics and Information Technology (CTIT)
Number of pages23
StatePublished - Feb 1999

Publication series

NameCTIT technical reports series
PublisherUniversity of Twente, Centre for Telematics and Information Technology (CTIT)
No.99-03

Fingerprint

Rare events
Simulation
Markov process
Model
Rare event simulation
Change of measure
Restart
Importance sampling
Sampling methods
Conditional distribution
Exceed

Keywords

  • METIS-118577
  • IR-18058
  • EWI-5975

Cite this

Garvels, M. J. J., & Kroese, D. (1999). On the entrance distribution in RESTART simulation. (CTIT technical reports series; No. 99-03). Enschede: Centre for Telematics and Information Technology (CTIT).

Garvels, M.J.J.; Kroese, Dirk / On the entrance distribution in RESTART simulation.

Enschede : Centre for Telematics and Information Technology (CTIT), 1999. 23 p. (CTIT technical reports series; No. 99-03).

Research output: ProfessionalReport

@book{4245360ada2c4891bf24361b49e1fd13,
title = "On the entrance distribution in RESTART simulation",
abstract = "The RESTART method is a widely applicable simulation technique for the estimation of rare event probabilities. The method is based on the idea to restart the simulation at certain intermediate stages, in order to generate more occurrences of the rare event. In many cases we are interested in the (rare) event that a real-valued function of some Markov process exceeds a high level. We explore the possibility of speeding up the RESTART method for such models by using certain conditional distributions of the Markov process, called 'entrance distributions'. As a by-product, we show, for certain models, a remarkable relationship between the entrance distribution and the optimal exponential change of measure used in the Importance Sampling method, the other main rare event simulation technique. We find that using the entrance distribution in the RESTART method will usually yield a significant reduction in simulation effort in situations where the method can be applied. The standard RESTART method is more robust, but is not as efficient as the new method since it does not use any information on the system we are simulating.",
keywords = "METIS-118577, IR-18058, EWI-5975",
author = "M.J.J. Garvels and Dirk Kroese",
note = "Imported from CTIT",
year = "1999",
month = "2",
series = "CTIT technical reports series",
publisher = "Centre for Telematics and Information Technology (CTIT)",
number = "99-03",
address = "Netherlands",

}

Garvels, MJJ & Kroese, D 1999, On the entrance distribution in RESTART simulation. CTIT technical reports series, no. 99-03, Centre for Telematics and Information Technology (CTIT), Enschede.

On the entrance distribution in RESTART simulation. / Garvels, M.J.J.; Kroese, Dirk.

Enschede : Centre for Telematics and Information Technology (CTIT), 1999. 23 p. (CTIT technical reports series; No. 99-03).

Research output: ProfessionalReport

TY - BOOK

T1 - On the entrance distribution in RESTART simulation

AU - Garvels,M.J.J.

AU - Kroese,Dirk

N1 - Imported from CTIT

PY - 1999/2

Y1 - 1999/2

N2 - The RESTART method is a widely applicable simulation technique for the estimation of rare event probabilities. The method is based on the idea to restart the simulation at certain intermediate stages, in order to generate more occurrences of the rare event. In many cases we are interested in the (rare) event that a real-valued function of some Markov process exceeds a high level. We explore the possibility of speeding up the RESTART method for such models by using certain conditional distributions of the Markov process, called 'entrance distributions'. As a by-product, we show, for certain models, a remarkable relationship between the entrance distribution and the optimal exponential change of measure used in the Importance Sampling method, the other main rare event simulation technique. We find that using the entrance distribution in the RESTART method will usually yield a significant reduction in simulation effort in situations where the method can be applied. The standard RESTART method is more robust, but is not as efficient as the new method since it does not use any information on the system we are simulating.

AB - The RESTART method is a widely applicable simulation technique for the estimation of rare event probabilities. The method is based on the idea to restart the simulation at certain intermediate stages, in order to generate more occurrences of the rare event. In many cases we are interested in the (rare) event that a real-valued function of some Markov process exceeds a high level. We explore the possibility of speeding up the RESTART method for such models by using certain conditional distributions of the Markov process, called 'entrance distributions'. As a by-product, we show, for certain models, a remarkable relationship between the entrance distribution and the optimal exponential change of measure used in the Importance Sampling method, the other main rare event simulation technique. We find that using the entrance distribution in the RESTART method will usually yield a significant reduction in simulation effort in situations where the method can be applied. The standard RESTART method is more robust, but is not as efficient as the new method since it does not use any information on the system we are simulating.

KW - METIS-118577

KW - IR-18058

KW - EWI-5975

M3 - Report

T3 - CTIT technical reports series

BT - On the entrance distribution in RESTART simulation

PB - Centre for Telematics and Information Technology (CTIT)

ER -

Garvels MJJ, Kroese D. On the entrance distribution in RESTART simulation. Enschede: Centre for Telematics and Information Technology (CTIT), 1999. 23 p. (CTIT technical reports series; 99-03).