Journal of Applied Mathematics and Stochastic Analysis
Volume 7 (1994), Issue 3, Pages 331-336
doi:10.1155/S1048953394000286
Abstract
A basic identity is proven and applied to obtain new simulation estimators
concerning (a) system reliability, (b) a multi-valued system. We show that the
variance of this new estimator is often of the order α2 when the usual raw
estimator has variance of the order α and α is small. We also indicate how this
estimator can be combined with standard variance reduction techniques of
antithetic variables, stratified sampling and importance sampling.