Textbook Notes (290,000)

CA (170,000)

York (10,000)

OMIS (50)

OMIS 2010 (30)

Alan Marshall (20)

Chapter 17

# OMIS 2010 Chapter Notes - Chapter 17: Response Surface Methodology, Linear Regression, F-Test

by OC214870

School

York UniversityDepartment

Operations Management and Information SystemCourse Code

OMIS 2010Professor

Alan MarshallChapter

17This

**preview**shows page 1. to view the full**4 pages of the document.**Chapter 17: Multiple Regression

Model and Required Conditions

-Assume that “k” independent variables potentially related to dependent variable

-Y = B0 + B1x1 + B2x2…..+ Bkxk +e

oY is the dependent variable

oX1, x2….. xk are independent variables

oE is error variable

oB0, B1…. Bk are coefficients

-Deviations between additional independent variables included, deviations between predicted values

of y” and actual values of “y” will occur

-Response surface: graphical depiction of equation when more than independent variable is in regression

model

Required Conditions for Error Variable

-Probability distribution of the error variable is normal

-Mean of error variable is 0

-Standard deviation of error variable is a constant

-Errors are independent

Estimating Coefficients and Assessing the Model

-Determine how well model fits the data

-If fit is poor then no point in further analysis of coefficients of that model

1) Select variables that you think are linearly related to dependent variable

2) Generate coefficients and stats used to assess the model

3) Diagnose violations of required conditions

4) Assess models fit; use standard error of estimate, coefficient of determination and F-test of

ANOVA

5) Interpret the coefficient; use the model to predict value of dependent variable or estimate expected

value of dependent variable

-Only screen independent variables and include those that affect dependent variable

###### You're Reading a Preview

Unlock to view full version