Basic Business Statistics

12th Edition

Chapter 14

Introduction to Multiple Regression

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-1
 Learning Objectives

In this chapter, you learn:

 How to develop a multiple regression model

 How to interpret the regression coefficients

 How to determine which independent variables to

include in the regression model
 How to determine which independent variables are more
important in predicting a dependent variable
 How to use categorical variables in a regression model

 How to predict a categorical dependent variable using

logistic regression

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-2
 The Multiple Regression
Model
DCOVA
Idea: Examine the linear relationship between
1 dependent (Y) & 2 or more independent variables (Xi)

Multiple Regression Model with k Independent Variables:

Y-intercept Population slopes Random Error

Y
i
 β 0  β 1 X 1i  β 2 X 2i
   βk X ki
 εi

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-3
 Multiple Regression Equation
DCOVA
The coefficients of the multiple regression model are
estimated using sample data

Multiple regression equation with k independent variables:

Estimated Estimated
(or predicted) Estimated slope coefficients
intercept
value of Y

Yˆ  b 0  b 1 X 1i  b 2 X    b k X
i 2i ki

In this chapter we will use Excel or Minitab to obtain the

regression slope coefficients and other regression
summary measures.
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-4
 Multiple Regression Equation
(continued)
Two variable model DCOVA
Y
Ŷ  b 0  b 1X 1  b 2 X 2

X 1

b le
ia
v ar
r
fo
l ope X2
S
a riabl e X2
f or v
S lope

X1
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-5
 Example:

2 Independent Variables
DCOVA
 A distributor of frozen dessert pies wants to
evaluate factors thought to influence demand
 Dependent variable: Pie sales (units per week)
 Independent variables: Price (in $)
Advertising ($100’s)

 Data are collected for 15 weeks

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-6
 Pie Sales Example
Pie Price Advertising DCOVA
Week Sales ($) ($100s) Multiple regression equation:
1 350 5.50 3.3
2 460 7.50 3.3
3 350 8.00 3.0 Sales = b0 + b1 (Price)
4
5
430
350
8.00
6.80
4.5
3.0
+ b2 (Advertising)
6 380 7.50 4.0
7 430 4.50 3.0
8 470 6.40 3.7
9 450 7.00 3.5
10 490 5.00 4.0
11 340 7.20 3.5
12 300 7.90 3.2
13 440 5.90 4.0
14 450 5.00 3.5
15 300 7.00 2.7

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-7
 Excel Multiple Regression Output
DCOVA
Regression Statistics
Multiple R 0.72213

R Square 0.52148
Adjusted R Square 0.44172
Standard Error 47.46341 Sales  306.526 - 24.975(Pri ce)  74.131(Adv ertising)
Observations 15

ANOVA df SS MS F Significance F
Regression 2 29460.027 14730.013 6.53861 0.01201
Residual 12 27033.306 2252.776
Total 14 56493.333

Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 306.52619 114.25389 2.68285 0.01993 57.58835 555.46404
Price -24.97509 10.83213 -2.30565 0.03979 -48.57626 -1.37392
Advertising 74.13096 25.96732 2.85478 0.01449 17.55303 130.70888

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-8
 Minitab Multiple Regression Output
DCOVA
Sales  306.526 - 24.975(Pri ce)  74.131(Adv ertising)

The regression equation is

Sales = 307 - 25.0 Price + 74.1 Advertising

Predictor Coef SE Coef T P

Constant 306.50 114.30 2.68 0.020
Price -24.98 10.83 -2.31 0.040
Advertising 74.13 25.97 2.85 0.014

S = 47.4634 R-Sq = 52.1% R-Sq(adj) = 44.2%

Analysis of Variance

Source DF SS MS F P
Regression 2 29460 14730 6.54 0.012
Residual Error 12 27033 2253
Total 14 56493

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-9
 The Multiple Regression Equation
DCOVA

Sales  306.526 - 24.975(Pri ce)  74.131(Adv ertising)

where
Sales is in number of pies per week
Price is in $
Advertising is in $100’s.
b1 = -24.975: sales b2 = 74.131: sales will
will decrease, on increase, on average,
average, by 24.975 by 74.131 pies per
pies per week for week for each $100
each $1 increase in increase in
selling price, net of advertising, net of the
the effects of effects of changes
changes due to due to price
advertising
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-10
 Using The Equation to Make
Predictions
DCOVA
Predict sales for a week in which the selling
price is $5.50 and advertising is $350:

Sales  306.526 - 24.975(Pri ce)  74.131(Adv ertising)

 306.526 - 24.975 (5.50)  74.131 (3.5)

 428.62

Note that Advertising is

Predicted sales in $100’s, so $350 means
that X2 = 3.5
is 428.62 pies

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-11
 Predictions in Excel using PHStat
DCOVA
 PHStat | regression | multiple regression …

Check the
“confidence and
prediction interval
estimates” box

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-12
 Predictions in PHStat
(continued)
DCOVA

Input values

<
Predicted Y value
Confidence interval for the
mean value of Y, given
these X values

Prediction interval for an

individual Y value, given
these X values
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-13
 Predictions in Minitab DCOVA
Confidence interval for
the mean value of Y,
given these X values
Predicted Values for New Observations

New
Obs Fit SE Fit 95% CI 95% PI
1 428.6 17.2 (391.1, 466.1) (318.6, 538.6)
Predicted Ŷ value

Values of Predictors for New Observations

New
Obs Price Advertising
1 5.50 3.50 Prediction interval
for an individual Y
value, given these X
Input values values

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-14
 Coefficient of

Multiple Determination
DCOVA
 Reports the proportion of total variation in Y
explained by all X variables taken together

SSR regression sum of squares

r
2
 
SST total sum of squares

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-15
 Multiple Coefficient of

Determination In Excel DCOVA
Regression Statistics
SSR 29460.0
Multiple R 0.72213
r
2
   .52148
R Square 0.52148 SST 56493.3
Adjusted R Square 0.44172
Standard Error 47.46341
52.1% of the variation in pie sales is
Observations 15
explained by the variation in price
and advertising
ANOVA df SS MS F Significance F
Regression 2 29460.027 14730.013 6.53861 0.01201
Residual 12 27033.306 2252.776
Total 14 56493.333

Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 306.52619 114.25389 2.68285 0.01993 57.58835 555.46404
Price -24.97509 10.83213 -2.30565 0.03979 -48.57626 -1.37392
Advertising 74.13096 25.96732 2.85478 0.01449 17.55303 130.70888

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-16
 Multiple Coefficient of

Determination In Minitab
DCOVA

The regression equation is

Sales = 307 - 25.0 Price + 74.1 Advertising

Predictor Coef SE Coef T P

SSR 29460.0
Constant 306.50 114.30 2.68 0.020
Price -24.98 10.83 -2.31 0.040
r
2
   .52148
SST 56493.3
Advertising 74.13 25.97 2.85 0.014

S = 47.4634 R-Sq = 52.1% R-Sq(adj) = 44.2%

Analysis of Variance
52.1% of the variation in pie sales
Source DF SS MS F P is explained by the variation in
Regression 2 29460 14730 6.54 0.012 price and advertising
Residual Error 12 27033 2253
Total 14 56493

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-17
 Adjusted r2
DCOVA
 r2 never decreases when a new X variable is
added to the model
 This can be a disadvantage when comparing

models
 What is the net effect of adding a new variable?
 We lose a degree of freedom when a new X

variable is added
 Did the new X variable add enough

explanatory power to offset the loss of one

degree of freedom?
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-18
 Adjusted r2 DCOVA
(continued)
 Shows the proportion of variation in Y explained
by all X variables adjusted for the number of X
variables used
 2  n  1 
r adj
2
 1  (1  r )  
  n  k  1 
(where n = sample size, k = number of independent variables)

 Penalize excessive use of unimportant independent

variables
 Smaller than r2
 Useful in comparing among models
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-19
 Adjusted r2 in Excel
DCOVA
Regression Statistics
Multiple R 0.72213
2
r adj  .44172
R Square 0.52148
Adjusted R Square 0.44172 44.2% of the variation in pie sales is explained
Standard Error 47.46341 by the variation in price and advertising, taking
Observations 15 into account the sample size and number of
independent variables
ANOVA df SS MS F Significance F
Regression 2 29460.027 14730.013 6.53861 0.01201
Residual 12 27033.306 2252.776
Total 14 56493.333

Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 306.52619 114.25389 2.68285 0.01993 57.58835 555.46404
Price -24.97509 10.83213 -2.30565 0.03979 -48.57626 -1.37392
Advertising 74.13096 25.96732 2.85478 0.01449 17.55303 130.70888

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-20
 Adjusted r 2 in Minitab
DCOVA

The regression equation is

2
r adj  .44172
Sales = 307 - 25.0 Price + 74.1 Advertising

Predictor Coef SE Coef T P

Constant 306.50 114.30 2.68 0.020
Price -24.98 10.83 -2.31 0.040
Advertising 74.13 25.97 2.85 0.014
44.2% of the variation in pie sales is
S = 47.4634 R-Sq = 52.1% R-Sq(adj) = 44.2% explained by the variation in price
and advertising, taking into account
Analysis of Variance
the sample size and number of
Source DF SS MS F P independent variables
Regression 2 29460 14730 6.54 0.012
Residual Error 12 27033 2253
Total 14 56493

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-21
 Is the Model Significant?
DCOVA
 F Test for Overall Significance of the Model
 Shows if there is a linear relationship between all
of the X variables considered together and Y
 Use F-test statistic
 Hypotheses:
H0: β1 = β2 = … = βk = 0 (no linear relationship)
H1: at least one βi ≠ 0 (at least one independent
variable affects Y)

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-22
 F Test for Overall Significance
DCOVA
 Test statistic:
SSR
MSR k
F STAT  
MSE SSE
n  k 1

where FSTAT has numerator d.f. = k and

denominator d.f. = (n – k - 1)

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-23
 F Test for Overall Significance In
Excel DCOVA
(continued)
Regression Statistics
Multiple R 0.72213
MSR 14730.0
R Square 0.52148
F STAT    6.5386
Adjusted R Square 0.44172 MSE 2252.8
Standard Error 47.46341
With 2 and 12 degrees P-value for
Observations 15
of freedom the F Test

ANOVA df SS MS F Significance F
Regression 2 29460.027 14730.013 6.53861 0.01201
Residual 12 27033.306 2252.776
Total 14 56493.333

Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 306.52619 114.25389 2.68285 0.01993 57.58835 555.46404
Price -24.97509 10.83213 -2.30565 0.03979 -48.57626 -1.37392
Advertising 74.13096 25.96732 2.85478 0.01449 17.55303 130.70888

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-24
 F Test for Overall Significance In
Minitab DCOVA

The regression equation is

Sales = 307 - 25.0 Price + 74.1 Advertising

Predictor Coef SE Coef T P

Constant 306.50 114.30 2.68 0.020
Price -24.98 10.83 -2.31 0.040 MSR 14730.0
Advertising 74.13 25.97 2.85 0.014 F STAT    6.5386
MSE 2252.8
S = 47.4634 R-Sq = 52.1% R-Sq(adj) = 44.2%

Analysis of Variance

Source DF SS MS F P
Regression 2 29460 14730 6.54 0.012
Residual Error 12 27033 2253
Total 14 56493

With 2 and 12 degrees P-value for

of freedom the F Test

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-25
 F Test for Overall Significance
(continued)

H0: β1 = β2 = 0 Test Statistic: DCOVA

H1: β1 and β2 not both zero MSR
F STAT   6.5386
 = .05 MSE
df1= 2 df2 = 12
Decision:
Critical Value: Since FSTAT test statistic is
F0.05 = 3.885 in the rejection region (p-
value < .05), reject H0
 = .05
Conclusion:
0 Do not Reject H0
F There is evidence that at least one
reject H0 independent variable affects Y
F0.05 = 3.885
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-26
 Residuals in Multiple Regression
DCOVA
Two variable model
Y Sample
Yi observation Ŷ  b 0  b 1X 1  b 2 X 2
Residual = ei
<

= (Yi – Yi)
<

Yi

x2i
X2
x1i
The best fit equation is found
X1 by minimizing the sum of
squared errors, e2
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-27
 Multiple Regression Assumptions
DCOVA
Errors (residuals) from the regression model:

<
ei = (Yi – Yi)

Assumptions:
 The errors are normally distributed

 Errors have a constant variance

 The model errors are independent

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-28
 Residual Plots Used

in Multiple Regression
DCOVA
 These residual plots are used in multiple
regression:

<
 Residuals vs. Yi
 Residuals vs. X1i
 Residuals vs. X2i
 Residuals vs. time (if time series data)

Use the residual plots to check for

violations of regression assumptions

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-29
 Are Individual Variables
Significant?
DCOVA


Use t tests of individual variable slopes
 Shows if there is a linear relationship between the
variable Xj and Y holding constant the effects of
other X variables


Hypotheses:
 H0: βj = 0 (no linear relationship)
 H1: βj ≠ 0 (linear relationship does exist
between Xj and Y)

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-30
 Are Individual Variables
Significant?
(continued)
DCOVA
H0: βj = 0 (no linear relationship)
H1: βj ≠ 0 (linear relationship does exist
between Xj and Y)

Test Statistic:

bj 0
t STAT  (df = n – k – 1)
Sb
j

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-31
 Are Individual Variables
Significant? Excel Output (continued)
DCOVA
Regression Statistics
Multiple R 0.72213
t Stat for Price is tSTAT = -2.306, with p-
R Square 0.52148
value .0398
Adjusted R Square 0.44172
Standard Error 47.46341 t Stat for Advertising is tSTAT = 2.855,
Observations 15 with p-value .0145

ANOVA df SS MS F Significance F
Regression 2 29460.027 14730.013 6.53861 0.01201
Residual 12 27033.306 2252.776
Total 14 56493.333

Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 306.52619 114.25389 2.68285 0.01993 57.58835 555.46404
Price -24.97509 10.83213 -2.30565 0.03979 -48.57626 -1.37392
Advertising 74.13096 25.96732 2.85478 0.01449 17.55303 130.70888

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-32
 Are Individual Variables Significant?
Minitab Output
DCOVA

The regression equation is

Sales = 307 - 25.0 Price + 74.1 Advertising

Predictor Coef SE Coef T P

Constant 306.50 114.30 2.68 0.020
Price -24.98 10.83 -2.31 0.040 t Stat for Price is tSTAT = -2.31, with p-
Advertising 74.13 25.97 2.85 0.014
value .040
S = 47.4634 R-Sq = 52.1% R-Sq(adj) = 44.2%

Analysis of Variance
t Stat for Advertising is tSTAT = 2.85, with
p-value .014
Source DF SS MS F P
Regression 2 29460 14730 6.54 0.012
Residual Error 12 27033 2253
Total 14 56493

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-33
 Inferences about the Slope:

t Test Example
DCOVA
From the Excel output:
H0: βj = 0
H1: βj  0 For Price tSTAT = -2.306, with p-value .0398

For Advertising tSTAT = 2.855, with p-value .0145

d.f. = 15-2-1 = 12
 = .05 The test statistic for each variable falls
t/2 = 2.1788 in the rejection region (p-values < .05)
Decision:
/2=.025 /2=.025 Reject H0 for each variable
Conclusion:
There is evidence that both
Reject H0
-tα/2
Do not reject H0
tα/2
Reject H0 Price and Advertising affect
0
-2.1788 2.1788 pie sales at  = .05
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-34
 Confidence Interval Estimate

for the Slope DCOVA
Confidence interval for the population slope βj

bj  tα /2 S b where t has
(n – k – 1) d.f.
j

Coefficients Standard Error

Intercept 306.52619 114.25389 Here, t has
Price -24.97509 10.83213
(15 – 2 – 1) = 12 d.f.
Advertising 74.13096 25.96732

Example: Form a 95% confidence interval for the effect of changes in price
(X1) on pie sales:
-24.975 ± (2.1788)(10.832)
So the interval is (-48.576 , -1.374)
(This interval does not contain zero, so price has a significant effect on sales)
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-35
 Confidence Interval Estimate

for the Slope DCOVA
(continued)
Confidence interval for the population slope βj

Coefficients Standard Error … Lower 95% Upper 95%

Intercept 306.52619 114.25389 … 57.58835 555.46404
Price -24.97509 10.83213 … -48.57626 -1.37392
Advertising 74.13096 25.96732 … 17.55303 130.70888

Example: Excel output also reports these interval endpoints:

Weekly sales are estimated to be reduced by between 1.37 to
48.58 pies for each increase of $1 in the selling price, holding the
effect of price constant

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-36
 Testing Portions of the
Multiple Regression Model
DCOVA
 Contribution of a Single Independent Variable Xj

SSR(Xj | all variables except Xj)

= SSR (all variables) – SSR(all variables except Xj)

 Measures the contribution of Xj in explaining the

total variation in Y (SST)

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-37
 Testing Portions of the
Multiple Regression Model
(continued)
DCOVA
Contribution of a Single Independent Variable Xj,
assuming all other variables are already included
(consider here a 2-variable model):

SSR(X1 | X2)
= SSR (all variables) – SSR(X2)

From ANOVA section of From ANOVA section of

regression for regression for
Yˆ  b 0  b1X 1  b 2 X Yˆ  b0  b2X
2 2

Measures the contribution of X1 in explaining SST

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-38
 The Partial F-Test Statistic
DCOVA
 Consider the hypothesis test:
H0: variable Xj does not significantly improve the model after all
other variables are included
H1: variable Xj significantly improves the model after all other
variables are included

 Test using the F-test statistic:

(with 1 and n-k-1 d.f.)

SSR (X j | all variables except j)

F STAT 
MSE
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-39
 Testing Portions of Model:
Example
DCOVA

Example: Frozen dessert pies

Test at the  = .05 level

to determine whether
the price variable
significantly improves
the model given that
advertising is included

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-40
 Testing Portions of Model:
Example
(continued)

H0: X1 (price) does not improve the model DCOVA

with X2 (advertising) included
H1: X1 does improve model

 = .05, df = 1 and 12
F0.05 = 4.75
(For X1 and X2) (For X2 only)
ANOVA ANOVA
df SS MS df SS
29460.0268 14730.0134 Regression 1 17484.22249
Regression 2 7 3
Residual 13 39009.11085
27033.3064 2252.77553
Total 14 56493.33333
Residual 12 7 9
56493.3333
Total 14 3
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-41
 Testing Portions of Model:
Example (continued)
DCOVA
(For X1 and X2) (For X2 only)
ANOVA ANOVA
df SS MS df SS
29460.0268 14730.0134 Regression 1 17484.22249
Regression 2 7 3
Residual 13 39009.11085
27033.3064 2252.77553
Total 14 56493.33333
Residual 12 7 9
56493.3333
Total 14 3
SSR (X 1 | X 2 ) 29 , 460 . 03  17 , 484 . 22
F STAT    5 . 316
MSE(all) 2252 . 78

Conclusion: Since FSTAT = 5.316 > F0.05 = 4.75 Reject H0;

Adding X1 does improve model

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-42
 Relationship Between Test
Statistics
DCOVA
 The partial F test statistic developed in this section and
the t test statistic are both used to determine the
contribution of an independent variable to a multiple
regression model.
 The hypothesis tests associated with these two
statistics always result in the same decision (that is, the
p-values are identical).

2
t STAT  F STAT

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-43
 Coefficient of Partial
Determination for k variable model
DCOVA
2
r Yj.(all variables except j)

SSR (X | all variables except j)


j

SST  SSR(all variables)  SSR(X j

| all variables except j)

 Measures the proportion of variation in the dependent

variable that is explained by Xj while controlling for
(holding constant) the other independent variables

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-44
 Coefficient of Partial
Determination in Excel
DCOVA
 Coefficients of Partial Determination can be
found using Excel:

 PHStat | regression | multiple regression …

 Check the “coefficient of partial determination” box
R e g r e s s io n A n a ly s is
C o e ffic ie n ts o f P a r tia l D e te r m in a tio n

In t e r m e d i a t e C a l c u l a t i o n s
S S R (X 1 , X 2 ) 2 9 4 6 0 .0 2 6 8 7
SST 5 6 4 9 3 .3 3 3 3 3
S S R (X 2 ) 1 7 4 8 4 .2 2 2 4 9 S S R (X 1 | X 2 ) 1 1 9 7 5 .8 0 4 3 8
S S R (X 1 ) 1 1 1 0 0 .4 3 8 0 3 S S R (X 2 | X 1 ) 1 8 3 5 9 .5 8 8 8 4

C o e ffic ie n ts
r 2 Y 1 .2 0 .3 0 7 0 0 0 1 8 8
r 2 Y 2 .1 0 .4 0 4 4 5 9 5 2 4

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-45
 Using Dummy Variables
DCOVA
 A dummy variable is a categorical independent
variable with two levels:
 yes or no, on or off, male or female
 coded as 0 or 1
 Assumes the slopes associated with numerical
independent variables do not change with the
value for the categorical variable
 If more than two levels, the number of dummy
variables needed is (number of levels - 1)

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-46
 Dummy-Variable Example

(with 2 Levels)
DCOVA

Ŷ  b0 b 1
X1 b 2
X 2

Let:
Y = pie sales
X1 = price
X2 = holiday (X2 = 1 if a holiday occurred during the week)
(X2 = 0 if there was no holiday that week)

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-47
 Dummy-Variable Example

(with 2 Levels) (continued)
DCOVA
Ŷ  b0 b 1
X1 b 2
(1)  (b 0  b 2 )  b 1
X1 Holiday

Ŷ  b0 b 1
X1 b 2
(0)  b0 b 1
X1 No Holiday

Different Same
intercept slope
Y (sales)
If H0: β2 = 0 is
b0 + b2 rejected, then
Holid
ay (X
b0 2 = 1)
“Holiday” has a
No H significant effect
o lid a
y (X on pie sales
2 = 0)

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall

X1 (Price) Chap 14-48
 Interpreting the Dummy Variable
Coefficient (with 2 Levels)DCOVA
Example: Sales  300 - 30(Price)  15(Holiday )

Sales: number of pies sold per week

Price: pie price in $
1 If a holiday occurred during the week
Holiday:
0 If no holiday occurred

b2 = 15: on average, sales were 15 pies greater in

weeks with a holiday than in weeks without a
holiday, given the same price

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-49
 Dummy-Variable Models

(more than 2 Levels) DCOVA
 The number of dummy variables is one less
than the number of levels
 Example:
Y = house price ; X1 = square feet

 If style of the house is also thought to matter:

Style = ranch, split level, colonial

Three levels, so two dummy

variables are needed

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-50
 Dummy-Variable Models

(more than 2 Levels) DCOVA
(continued)

 Example: Let “colonial” be the default category, and let

X2 and X3 be used for the other two categories:

Y = house price
X1 = square feet
X2 = 1 if ranch, 0 otherwise
X3 = 1 if split level, 0 otherwise

The multiple regression equation is:

Ŷ  b 0  b 1X 1  b 2 X 2  b 3 X 3
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-51
 Interpreting the Dummy Variable
Coefficients (with 3 Levels)
DCOVA
Consider the regression equation:
Ŷ  20.43  0.045X 1
 23.53X 2
 18.84X 3

For a colonial: X2 = X3 = 0
With the same square feet, a
Ŷ  20.43  0.045X 1
ranch will have an estimated
average price of 23.53
For a ranch: X2 = 1; X3 = 0 thousand dollars more than a
colonial.
Ŷ  20.43  0.045X 1
 23.53
With the same square feet, a
For a split level: X2 = 0; X3 = 1 split-level will have an
estimated average price of
Ŷ  20.43  0.045X 1
 18.84 18.84 thousand dollars more
than a colonial.
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-52
 Interaction Between
Independent Variables
DCOVA
 Hypothesizes interaction between pairs of X
variables
 Response to one X variable may vary at different
levels of another X variable

 Contains two-way cross product terms


Ŷ  b 0  b 1X 1  b 2 X 2  b 3 X 3

 b 0  b 1 X 1  b 2 X 2  b 3 (X 1 X 2 )
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-53
 Effect of Interaction
DCOVA
 Given: Y  β 0  β 1X 1  β 2 X 2  β 3 X 1X 2  ε

 Without interaction term, effect of X1 on Y is

measured by β1
 With interaction term, effect of X1 on Y is
measured by β1 + β3 X2
 Effect changes as X2 changes

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-54
 Interaction Example
DCOVA
Suppose X2 is a dummy variable and the estimated
regression equation is Ŷ = 1 + 2X1 + 3X2 + 4X1X2
Y
12
X2 = 1:
8 Y = 1 + 2X1 + 3(1) + 4X1(1) = 4 + 6X1

4
X2 = 0:
Y = 1 + 2X1 + 3(0) + 4X1(0) = 1 + 2X1
0
X1
0 0.5 1 1.5
Slopes are different if the effect of X1 on Y depends on X2 value
Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-55
 Significance of Interaction Term
DCOVA
 Can perform a partial F test for the contribution
of a variable to see if the addition of an
interaction term improves the model

 Multiple interaction terms can be included

 Use a partial F test for the simultaneous contribution
of multiple variables to the model

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-56
 Simultaneous Contribution of
Independent Variables
DCOVA
 Use partial F test for the simultaneous
contribution of multiple variables to the model
 Let m variables be an additional set of variables
added simultaneously
 To test the hypothesis that the set of m variables
improves the model:

[SSR(all)  SSR (all except new set of m variables )] / m

F STAT 
MSE(all)

(where FSTAT has m and n-k-1 d.f.)

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-57
 Logistic Regression
DCOVA
 Used when the dependent variable Y is binary
(i.e., Y takes on only two values)
 Examples
 Customer prefers Brand A or Brand B
 Employee chooses to work full-time or part-time
 Loan is delinquent or is not delinquent
 Person voted in last election or did not
 Logistic regression allows you to predict the
probability of a particular categorical response

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-58
 Logistic Regression DCOVA
(continued)

 Logistic regression is based on the odds ratio,

which represents the probability of a success
compared with the probability of failure

probabilit y of success
Odds ratio 
1  probabilit y of success

 The logistic regression model is based on the

natural log of this odds ratio

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-59
 Logistic Regression
(continued)
DCOVA
Logistic Regression Model:

ln(odds ratio)  β 0  β 1 X 1i  β 2 X 2i
   βk X ki
 εi

Where k = number of independent variables in the model

εi = random error in observation i

Logistic Regression Equation:

ln(estimat ed odds ratio)  b 0  b 1 X 1i  b 2 X 2i

   b k X ki

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-60
 Estimated Odds Ratio and
Probability of Success
DCOVA

 Once you have the logistic regression equation,

compute the estimated odds ratio:

Estimated odds ratio e ln(estimat ed odds ratio)

 The estimated probability of success is

estimated odds ratio
Estimated probabilit y of success 
1  estimated odds ratio

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-61
 Chapter Summary
 Developed the multiple regression model
 Tested the significance of the multiple regression
model
 Discussed adjusted r2
 Discussed using residual plots to check model
assumptions
 Tested individual regression coefficients
 Tested portions of the regression model
 Used dummy variables
 Evaluated interaction effects
 Discussed logistic regression

Copyright ©2012 Pearson Education, Inc. publishing as Prentice Hall Chap 14-62