Common random numbers in multivariate simulations (Q2640337)

Common random numbers (CRN) is a reduction method used in simulation which has the purpose of reducing the variance of the estimator of a system parameter without increasing the bias. The point is that the systems should be compared under similar experimental conditions so that any observed differences will be due to differences in the structures. The aim of this paper is to generalize the technique of CRN for use in multivariate simulations. The conclusion is that the use of CRN can lead to a substantial reduction in the size of a confidence region and this reduction is the greatest one when the covariance structure is the same for both systems.