Professional Documents
Culture Documents
Multiple Regression
Multiple Regression Overview
• What is it?
• Why use it?
Multiple Regression
X3
Y
X1
X2
Example of a “Business-to-Business”
Application of Multiple Regression
Performance Measures
• Sources of variables?
• Number of independent variables?
Is More Better ?
• Method of variable entry?
• Interpretation of regression
functions?
Regression Variables?
Sources :
Prior Research ?
Practice – Current Business Situation ?
Theoretical Model ?
Intuition ?
Regression Variables ?
Issues :
Measurement Error – Both Dependent
& Independents.
Specification Error – Independents only.
•Too Few:
lowers prediction.
introduces bias.
•Too Many:
Reduces parsimony.
Masks effects of other variables.
Variable Selection Approaches:
• Confirmatory (Simultaneous).
• Sequential Search Methods:
Stepwise (variables not removed once included
in regression equation).
Forward Inclusion & Backward Elimination.
• Combinatorial (All-Possible-Subsets).
Interpretation of
Regression Results:
• Coefficient of Determination.
• Regression Coefficients
(Unstandardized – bivariate).
• Beta Coefficients (Standardized).
• Variables Entered.
• Multicollinearity ??
Statistical vs. Practical Significance ?
The F statistic is used to determine if the overall regression model
is statistically significant. If the F statistic is significant, it means it is
unlikely your sample will produce a large R2 when the population R2 is
actually zero. To be considered statistically significant, a rule of thumb
is there must be <.05 probability the results are due to chance.
Multicollinearity
Dummy Variables
Assumptions
Multicollinearity
. . . . is the correlation
among the
independent variables.
Multicollinearity Diagnostics:
. . . . a nonmetric independent
variable that has two (or more)
distinct levels, that are coded
0 and 1.
Selected Variables from Employee Survey
Dependent Variable
15. I am proud to tell others that I
work for Samouel’s restaurant. Strongly Strongly
Disagree Agree
1 2 3 4 5 6 7
Using SPSS to Examine Dummy Variables:
Y Deviation not
explained by
regression
Total Deviation
Y = average
Deviation
explained by
regression
X
Residuals Plots
Histogram
Dependent Variable: X15 -- Proud
10
4
Frequency
distribution (observed) is
shown as deviations from
the straight line. .75
Expected Cum Prob
.50
.25
0.00
0.00 .25 .50 .75 1.00
-1
-2
-2 -1 0 1 2 3