Author:
Winfried Grassmann
Affiliation:
University of Saskatchewan, Canada
Keyword(s):
Steady-state Simulation, Multiple Runs, Initialization Bias.
Related
Ontology
Subjects/Areas/Topics:
Computer Simulation Techniques
;
Simulation and Modeling
;
Simulation Tools and Platforms
Abstract:
In complex stochastic systems, Monte-Carlo simulation is often the only way to estimate equilibrium expectations.
The question then arises what is better: a single run of length T, or n runs, each of length T/n. In this
paper, it is argued that if there is a good state to start the simulation in, multiple runs may be advantageous.
To illustrate this, we use numerical examples. These examples are obtained by using deterministic methods,
that is, methods based on probability theory not using Monte-Carlo methods. The results of our numerical
calculations forced us to make a sharp distinction between the time to reach equilibrium and the appropriate
length of the warm-up period, and this distinguishes our study from earlier investigations.