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
Hi-index | 0.00 |
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.