Estimating Time Averages via Randomly-Spaced Observations
B. L. Fox and P. W. Glynn
SIAM Journal of Applied Mathematics, Vol. 47, 186-200 (1987)
To estimate continuous-time averages via randomly-spaced observations of discrete-event systems, we develop a point-process framework and use it to generalize both regenerative and stationaryprocess oriented simulation methodologies. We give consistent estimators, central limit theorems, and an effective bias-reducing jackknife. The impact on indirect estimation of transaction (customer) averages is discussed.