On the approximation of multiple integrals using multivariate ranked simulated sampling

Abstract

The idea of using ranked simulated sampling approach (RSIS) to improve the well known Monte Carlo methods of integration, introduced by Samawi [H.M. Samawi, More efficient Monte Carlo methods obtained by using ranked set simulated samples, Commun. Stat. Simulat. 28 (3) (1999) 699–713], is extended to multivariate ranked simulated sampling approach (MVRSIS) for multiple integration problems. It is demonstrated that this approach provides unbiased estimators and improves the performance of some of the Monte Carlo methods of multiple integrals approximation. This, results in large saving in terms of cost and time needed to attain a certain level of accuracy. Two illustrations using simulation are used to compare the relative performance of this approach relative to multivariate uniform simulation.