Statistics and deterministic simulation models: why not?

Author

Kleijnen, Jack P C

Author_Institution

Katholieke Univ. Brabant, Tilburg, Netherlands

fYear

1990

fDate

9-12 Dec 1990

Firstpage

344

Lastpage

346

Abstract

Deterministic simulation models are compared with random simulation models and real-life experiments. In deterministic simulation, no mathematical statistics is needed in the experimental design and in the least squares curve fitting. Further analysis, however, becomes possible if certain statistical models are specified for the fitting errors. Normally identically and independently distributed errors were proposed by the author in 1987. J. Sacks et al. (1989) proposed dependent errors with a specific correlation structure. Needs for further research are indicated

Keywords

curve fitting; design engineering; error analysis; least squares approximations; simulation; statistics; correlation structure; dependent errors; deterministic simulation models; experimental design; fitting errors; independently distributed errors; least squares curve fitting; random simulation models; real-life experiments; statistical models; Analytical models; Biological system modeling; Curve fitting; Design for experiments; Input variables; Least squares methods; Lifting equipment; Monte Carlo methods; Statistical distributions; Statistics;

fLanguage

English

Publisher

ieee

Conference_Titel

Simulation Conference, 1990. Proceedings., Winter

Conference_Location

New Orleans, LA

Print_ISBN

0-911801-72-3

Type

conf

DOI

10.1109/WSC.1990.129538

Filename

129538