Estimation of rare event probabilities using cross-entropy

Tito Homem-de-Mello, Reuven Y. Rubinstein

Resultado de la investigación: Contribución a una revistaArtículo de la conferenciarevisión exhaustiva

32 Citas (Scopus)

Resumen

This paper deals with estimation of probabilities of rare events in static simulation models using a fast adaptive two-stage procedure based on importance sampling and Kullback-Liebler's cross-entropy (CE). More specifically, at the first stage we estimate the optimal parameter vector in the importance sampling distribution using CE, and at the second stage we estimate the desired rare event probability using importance sampling (likelihood ratios). Some theoretical aspects of the proposed method, including its convergence, are established. The numerical results presented suggest that the method effectively estimates rare event probabilities.

Idioma originalInglés
Páginas (desde-hasta)310-319
Número de páginas10
PublicaciónWinter Simulation Conference Proceedings
Volumen1
EstadoPublicada - 2002
Publicado de forma externa
EventoProceedings of the 2002 Winter Simulation Conference - San Diego, CA, Estados Unidos
Duración: 8 dic. 200211 dic. 2002

Huella

Profundice en los temas de investigación de 'Estimation of rare event probabilities using cross-entropy'. En conjunto forman una huella única.

Citar esto