The effect of correlation on ordinal comparison of discrete event dynamic systems

Research has demonstrated that ordinal comparison, i.e. comparing relative orders of performance measures, converges much faster than the performance measures themselves do. However, the actual rate is greatly influenced by the correlation among systems under different designs. We investigate the effect of correlation on ordinal comparison of discrete event dynamic systems. It is shown that positive quadrant dependence increases the rate of convergence for ordinal comparison while negative quadrant dependence decreases the rate. The results of this paper also show that the rate is maximized by using the scheme of common random numbers, a widely-used technique for variance reduction.