Some Confidence Regions for Traffic Intensity Vector.

Using the Consistent and Asymptotically Normal(CAN) estimator and its covariance matrix (A;) 100(1 - α)% confidence region for traffic intensity vector P with no assumption of arrival and service time distribution is constructed in this paper. Also Standard Bootstrap (SB), Bayesian Bootstrap(BB) and percentile bootstrap (PB) are applied to develop the confidence regions for traffic intensity vector P with confidence level 100(1 - α)%: Simulation study is undertaken to evaluate the performances of the confidence regions in terms of their coverage area percentage, average area and relative coverage area. Calibration technique is used to improve the coverage area percentages of confidence regions. [ABSTRACT FROM AUTHOR]