A regression technique for determining steady state conditions in time series simulations

Authors:
Charles W. Beall
Affiliations:
University of Hawaii, Honolulu, Hawaii
Venue:
WSC '82 Proceedings of the 14th conference on Winter Simulation - Volume 2
Year:
1982

Citing 3
Cited 1

Principles of Discrete Event Simulation

Principles of Discrete Event Simulation
System Simulation

System Simulation
Time Series Analysis, Forecasting and Control

Time Series Analysis, Forecasting and Control

Steady-state simulation of queueing processes: survey of problems and solutions

ACM Computing Surveys (CSUR)

Quantified Score

Hi-index	0.00

Visualization

Abstract

In many situations, the simulator finds himself faced with the problem of simulating a time series whose initial state is different from the state of the system after a large number of observations have been made. This is frequently referred to as the problem of the initial bias or the initial tranient problem. In this paper, we borrow from the theory of convergence in distribution to develop a criterion for convergence to a steady state process and then develop estimation techniques based on simple linear regression to determine when the time series under investigation has converged to a steady state process.