# Regression Models and Experimental Designs: A Tutorial for Simulation Analaysts

Research output: Working paperDiscussion paperOther research output

### Abstract

This tutorial explains the basics of linear regression models. especially low-order polynomials. and the corresponding statistical designs. namely, designs of resolution III, IV, V, and Central Composite Designs (CCDs).This tutorial assumes 'white noise', which means that the residuals of the fitted linear regression model are normally, independently, and identically distributed with zero mean.The tutorial gathers statistical results that are scattered throughout the literature on mathematical statistics, and presents these results in a form that is understandable to simulation analysts.
Original language English Tilburg Operations research 31 2006-10 Published - 2006

### Publication series

Name CentER Discussion Paper 2006-10

### Fingerprint

Experimental design
Regression Model
Linear Regression Model
Simulation
White noise
Identically distributed
Composite
Statistics
Polynomial
Zero
Design

### Keywords

• metamodels
• fractional factorial designs
• Plackett-Burman designs
• factor interactions
• validation
• cross-validation

### Cite this

Kleijnen, J. P. C. (2006). Regression Models and Experimental Designs: A Tutorial for Simulation Analaysts. (CentER Discussion Paper; Vol. 2006-10). Tilburg: Operations research.
Kleijnen, J.P.C. / Regression Models and Experimental Designs : A Tutorial for Simulation Analaysts. Tilburg : Operations research, 2006. (CentER Discussion Paper).
@techreport{7b8ecddbf49e4b80865baa8c76d813fa,
title = "Regression Models and Experimental Designs: A Tutorial for Simulation Analaysts",
abstract = "This tutorial explains the basics of linear regression models. especially low-order polynomials. and the corresponding statistical designs. namely, designs of resolution III, IV, V, and Central Composite Designs (CCDs).This tutorial assumes 'white noise', which means that the residuals of the fitted linear regression model are normally, independently, and identically distributed with zero mean.The tutorial gathers statistical results that are scattered throughout the literature on mathematical statistics, and presents these results in a form that is understandable to simulation analysts.",
keywords = "metamodels, fractional factorial designs, Plackett-Burman designs, factor interactions, validation, cross-validation",
author = "J.P.C. Kleijnen",
note = "Subsequently published in Proceedings of the Winter Simulation Conference 2007 Pagination: 31",
year = "2006",
language = "English",
volume = "2006-10",
series = "CentER Discussion Paper",
publisher = "Operations research",
type = "WorkingPaper",
institution = "Operations research",

}

Kleijnen, JPC 2006 'Regression Models and Experimental Designs: A Tutorial for Simulation Analaysts' CentER Discussion Paper, vol. 2006-10, Operations research, Tilburg.
Tilburg : Operations research, 2006. (CentER Discussion Paper; Vol. 2006-10).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Regression Models and Experimental Designs

T2 - A Tutorial for Simulation Analaysts

AU - Kleijnen, J.P.C.

N1 - Subsequently published in Proceedings of the Winter Simulation Conference 2007 Pagination: 31

PY - 2006

Y1 - 2006

N2 - This tutorial explains the basics of linear regression models. especially low-order polynomials. and the corresponding statistical designs. namely, designs of resolution III, IV, V, and Central Composite Designs (CCDs).This tutorial assumes 'white noise', which means that the residuals of the fitted linear regression model are normally, independently, and identically distributed with zero mean.The tutorial gathers statistical results that are scattered throughout the literature on mathematical statistics, and presents these results in a form that is understandable to simulation analysts.

AB - This tutorial explains the basics of linear regression models. especially low-order polynomials. and the corresponding statistical designs. namely, designs of resolution III, IV, V, and Central Composite Designs (CCDs).This tutorial assumes 'white noise', which means that the residuals of the fitted linear regression model are normally, independently, and identically distributed with zero mean.The tutorial gathers statistical results that are scattered throughout the literature on mathematical statistics, and presents these results in a form that is understandable to simulation analysts.

KW - metamodels

KW - fractional factorial designs

KW - Plackett-Burman designs

KW - factor interactions

KW - validation

KW - cross-validation

M3 - Discussion paper

VL - 2006-10

T3 - CentER Discussion Paper

BT - Regression Models and Experimental Designs

PB - Operations research

CY - Tilburg

ER -

Kleijnen JPC. Regression Models and Experimental Designs: A Tutorial for Simulation Analaysts. Tilburg: Operations research. 2006. (CentER Discussion Paper).