Initial bias and estimation error in discrete event simulation

Authors:
W. K. Grassmann
Affiliations:
University of Saskatchewan, Saskatoon, Saskatchewan
Venue:
WSC '82 Proceedings of the 14th conference on Winter Simulation - Volume 2
Year:
1982

Citing 0
Cited 4

Computation of the Asymptotic Bias and Variance for Simulation of Markov Reward Models

SS '96 Proceedings of the 29th Annual Simulation Symposium (SS '96)
Warm-up periods in simulation can be detrimental

Probability in the Engineering and Informational Sciences
Rethinking the initialization bias problem in steady-state discrete event simulation

Proceedings of the Winter Simulation Conference
Factors affecting warm-up periods in discrete event simulation

Simulation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper analyses several stochastic systems and obtains expressions for the initial bias. The bias is compared with the estimation error of the same system starting in statistical equilibrium. It turns out that the bias and the estimation error are closely related to each other. Furthermore, it is shown that if the estimation error is small, the bias becomes negligible.