Smoothing the Time Series for Input and Output Analysis in System Simulation Experiments
Loading...
Authors
Lewis, Peter A.W.
Stevens, Jame G.
Subjects
Advisors
Date of Issue
1990
Date
Publisher
Language
Abstract
Classical methods of studying the behaviour of the output of a simulation model as a function of parameters (independent variables, factors, predictor variables) can be divided into global regression and smoothing (local regression). Neither of these methods are adequate, especially when the observation are a function of a time evolution variable and are probably highly correlated. Several new smoothing methods have been proposed recently for this problem, most of them based on the use of splines. This paper concentrates on the use of multivariate adaptive regression spline (MARS) methodology for this smoothing and characterization problem, and on the use of this methodology when there is serial correlation in the data so that lagged values of the observations can be used for predictor variables. The methodology is also useful when analyzing inputs to queues.
Type
Conference Paper
Description
Winter Simulation Conference
Series/Report No
Department
Department of Operations Research
Organization
Identifiers
NPS Report Number
Sponsors
Funder
Format
Citation
Distribution Statement
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.