ENGDAT2 Statistical Analysis for

Industrial Engineers

Introduction to Multiple Regression

De La Salle University :: Dept of Industrial Engineering

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
 THE MULTIPLE
REGRESSION MODEL
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

Yi  β 0  β1 X1i  β 2 X 2i      β k X ki  ε i
 MULTIPLE REGRESSION
EQUATION
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
value of Y intercept

ˆ  b  b X  b X    b X
Yi 0 1 1i 2 2i k ki
In this chapter we will use Excel or Minitab to obtain the
regression slope coefficients and other regression
summary measures.
 MULTIPLE REGRESSION
EQUATION
Two variable model
(continued)

Y
Ŷ  b0  b1X1  b 2 X 2

X1
e
abl
ri
r va
fo
l ope X2
S
f or v ariable X 2
Slope

X1
 EXAMPLE:
2 INDEPENDENT VARIABLES
 A distributor of frozen dessert pies wants to evaluate factors thought to
influence demand

 Dependent variable: Pie sales (units per week)

 Two Independent Price (in $)

variables: Advertising ($100’s)

Data are collected for 15 weeks

PIE SALES EXAMPLE
Pie Price Advertising
Week Sales ($) ($100s) Multiple regression equation:
1 350 5.50 3.3
2
3
460
350
7.50
8.00
3.3
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
 EXCEL MULTIPLE REGRESSION
OUTPUT
Regression Statistics
Multiple R 0.72213

R Square 0.52148
Adjusted R Square 0.44172
Standard Error 47.46341 Sales  306.526- 24.975(Price)  74.131(Advertising)
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
 MINITAB MULTIPLE REGRESSION
OUTPUT
Sales  306.526- 24.975(Price)  74.131(Advertising)

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
 THE MULTIPLE REGRESSION
EQUATION

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 changes effects of changes
due to advertising due to price
 USING THE EQUATION TO
MAKE PREDICTIONS
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
 PREDICTIONS IN MINITAB
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
ˆ value 1 428.6 17.2 (391.1, 466.1) (318.6, 538.6)
Predicted Y

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

Chap 14-12
 COEFFICIENT OF
MULTIPLE DETERMINATION

 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
 MULTIPLE COEFFICIENT OF
DETERMINATION IN EXCEL
Regression Statistics
Multiple R 0.72213 SSR 29460.0
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
Observations 15 is 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
 MULTIPLE COEFFICIENT OF
DETERMINATION IN MINITAB
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 SSR 29460.0
Price -24.98 10.83 -2.31 0.040
r2    .52148
Advertising 74.13 25.97 2.85 0.014
SST 56493.3

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

Analysis of Variance
52.1% of the variation in pie
Source DF SS MS F P sales is explained by the
Regression 2 29460 14730 6.54 0.012 variation in price and
Residual Error 12 27033 2253
Total 14 56493
advertising
ADJUSTED R2
 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?
ADJUSTED R2 (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 2
adj  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
 ADJUSTED R2 IN EXCEL
Regression Statistics
Multiple R 0.72213
2
radj  .44172
R Square 0.52148
Adjusted R Square 0.44172 44.2% of the variation in pie sales is
Standard Error 47.46341 explained by the variation in price and
Observations 15 advertising, taking 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
 ADJUSTED R2 IN MINITAB
The regression equation is
2
radj  .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
S = 47.4634 R-Sq = 52.1% R-Sq(adj) = 44.2% sales is explained by the
variation in price and
Analysis of Variance
advertising, taking into account
Source DF SS MS F P the sample size and number of
Regression 2 29460 14730 6.54 0.012 independent variables
Residual Error 12 27033 2253
Total 14 56493
IS THE MODEL SIGNIFICANT?
 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)
F TEST FOR OVERALL
SIGNIFICANCE
 Test statistic:

SSR
MSR k
FSTAT  
MSE SSE
n  k 1
where FSTAT has numerator d.f. = k and
denominator d.f. = (n – k - 1)
 F TEST FOR OVERALL
SIGNIFICANCE IN EXCEL
Regression Statistics
(continued)

Multiple R 0.72213

R Square 0.52148 MSR 14730.0

FSTAT    6.5386
Adjusted R Square 0.44172 MSE 2252.8
Standard Error 47.46341
Observations 15
With 2 and 12 degrees P-value for
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
F TEST FOR OVERALL
SIGNIFICANCE IN MINITAB
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 FSTAT    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
 F TEST FOR OVERALL
SIGNIFICANCE (continued)

H0: β1 = β2 = 0 Test Statistic:

H1: β1 and β2 not both zero MSR
FSTAT   6.5386
 = .05 MSE
df1= 2 df2 = 12 Decision:
Critical Since FSTAT test statistic is
Value:
in the rejection region (p-
F0.05 = 3.885
value < .05), reject H0
 = .05
Conclusion:
0 F There is evidence that at least one
Do not Reject H0
reject H0 independent variable affects Y
F0.05 = 3.885
 RESIDUALS IN MULTIPLE
REGRESSION
Two variable model
Y Sample
Yi observation Ŷ  b0  b1X1  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
 MULTIPLE REGRESSION
ASSUMPTIONS
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
 RESIDUAL PLOTS USED
IN MULTIPLE REGRESSION

 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
 ARE INDIVIDUAL VARIABLES
SIGNIFICANT?
 Use t tests of individual variable slopes
 Shows if there is a linear relationship between the variable X j 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)
ARE INDIVIDUAL VARIABLES
SIGNIFICANT?
(continued)

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
 ARE INDIVIDUAL VARIABLES
SIGNIFICANT? EXCEL OUTPUT
(continued)
Regression Statistics
Multiple R 0.72213
t Stat for Price is tSTAT = -2.306, with
R Square 0.52148 p-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
 ARE INDIVIDUAL VARIABLES
SIGNIFICANT? MINITAB OUTPUT
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 = -2.306, with
Advertising 74.13 25.97 2.85 0.014
p-value .0398
S = 47.4634 R-Sq = 52.1% R-Sq(adj) = 44.2%

Analysis of Variance
t Stat for Advertising is tSTAT = 2.855,
with p-value .0145
Source DF SS MS F P
Regression 2 29460 14730 6.54 0.012
Residual Error 12 27033 2253
Total 14 56493
 INFERENCES ABOUT THE SLOPE:
T TEST EXAMPLE

H0: βj = 0 From the Excel and Minitab output:

For Price tSTAT = -2.306, with p-value .0398
H1: βj  0
For Advertising tSTAT = 2.855, with p-value .0145
d.f. = 15-2-1 = 12
a = .05 The test statistic for each variable falls
t/2 = 2.1788 in the rejection region (p-values < .05)
Decision:
a/2=.025 a/2=.025 Reject H0 for each variable
Conclusion:
There is evidence that both
Reject H0 Do not reject H0
-tα/2 tα/2
Reject H0
Price and Advertising affect
0
-2.1788 2.1788 pie sales at  = .05
 CONFIDENCE INTERVAL ESTIMATE
FOR THE SLOPE
Confidence interval for the population slope βj

b j  tα / 2 Sb 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)
 CONFIDENCE INTERVAL ESTIMATE
FOR THE SLOPE
(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
 TESTING PORTIONS OF THE
MULTIPLE REGRESSION MODEL
 Contribution of a Single Independent Variable Xj

SSR(Xj | all variables except Xj)

= SSR (all variables) – SSR(all variables except X j)

 Measures the contribution of Xj in explaining the total variation in Y (SST)

 TESTING PORTIONS OF THE
MULTIPLE REGRESSION MODEL
(continued)
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
ˆ b b X b X
Y ˆ b b X
Y
0 1 1 2 2 0 2 2

Measures the contribution of X1 in explaining SST

THE PARTIAL F-TEST
STATISTIC
 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 variablesexcept j)
FSTAT 
MSE
TESTING PORTIONS OF
MODEL: EXAMPLE

Example: Frozen dessert pies

Test at the  = .05 level

to determine whether
the price variable
significantly improves
the model given that
advertising is included
 TESTING PORTIONS OF
MODEL: EXAMPLE
(continued)

H0: X1 (price) does not improve the model

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
Regression 2 29460.02687 14730.01343 Regression 1 17484.22249
Residual 12 27033.30647 2252.775539 Residual 13 39009.11085
Total 14 56493.33333 Total 14 56493.33333
 TESTING PORTIONS OF
MODEL: EXAMPLE
(continued)
(For X1 and X2) (For X2 only)
ANOVA ANOVA
df SS MS df SS
Regression 2 29460.02687 14730.01343 Regression 1 17484.22249
Residual 12 27033.30647 2252.775539 Residual 13 39009.11085
Total 14 56493.33333 Total 14 56493.33333

SSR (X1 | X 2 ) 29,460.03  17,484.22

FSTAT    5.316
MSE(all) 2252.78

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

Adding X1 does improve model
 RELATIONSHIP BETWEEN TEST
STATISTICS

 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
ta  F1,a
Where a = degrees of freedom

Chap 14-41
COEFFICIENT OF PARTIAL DETERMINATION
FOR K VARIABLE MODEL

2
rYj.(all variables except j)

SSR (X j | all variables except 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
USING DUMMY VARIABLES
 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)
 DUMMY-VARIABLE EXAMPLE
(WITH 2 LEVELS)

Ŷ  b0  b1 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)
 DUMMY-VARIABLE EXAMPLE
(WITH 2 LEVELS)
(continued)

Ŷ  b0  b1 X1  b 2 (1)  (b 0  b 2 )  b1 X1 Holiday

Ŷ  b0  b1 X1  b 2 (0)  b0  b1 X1 No Holiday

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

X1 (Price)
 INTERPRETING THE DUMMY VARIABLE
COEFFICIENT (WITH 2 LEVELS)

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
 DUMMY-VARIABLE MODELS
(MORE THAN 2 LEVELS)
 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
 DUMMY-VARIABLE MODELS
(MORE THAN 2 LEVELS)
(continued)
 Example: Let “colonial” be the default category, and let X 2 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  b0  b1X1  b 2 X 2  b3 X3
 INTERPRETING THE DUMMY VARIABLE
COEFFICIENTS (WITH 3 LEVELS)

Consider the regression equation:

Ŷ  20.43  0.045X1  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.
 INTERACTION BETWEEN
INDEPENDENT VARIABLES

 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  b0  b1X1  b 2 X 2  b3 X3

 b0  b1X1  b 2 X 2  b3 (X1X 2 )
EFFECT OF INTERACTION
 Given:
Y  β0  β1X1  β 2 X 2  β3 X1X 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

 INTERACTION EXAMPLE
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
 SIGNIFICANCE OF
INTERACTION TERM

 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
 SIMULTANEOUS CONTRIBUTION
OF INDEPENDENT VARIABLES

 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

FSTAT 
MSE(all)

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

 LOGISTIC REGRESSION

 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
 LOGISTIC REGRESSION
(continued)

 Logistic regression is based on the odds ratio, which represents the

probability of a success compared with the probability of failure

probability of success
Odds ratio 
1 probability of success
 The logistic regression model is based on the natural log of this odds
ratio
 LOGISTIC REGRESSION
(continued)

Logistic Regression Model:

ln(oddsratio)  β 0  β1X1i  β 2 X 2i      β k X ki  ε i

Where k = number of independent variables in the model

εi = random error in observation i

Logistic Regression Equation:

ln(estimated odds ratio)  b 0  b1 X1i  b 2 X 2i      b k X ki

 ESTIMATED ODDS RATIO AND
PROBABILITY OF SUCCESS

 Once you have the logistic regression equation, compute the estimated
odds ratio:

Estimated odds ratio  eln(estimat ed odds ratio)

 The estimated probability of success is

estimated odds ratio

Estimated probability of success 
1 estimated odds ratio
 MODEL BUILDING VIA
BACKWARD ELIMINATION OF
X VARIABLES
 From a set of possible independent variables Xi that can be associated with
a response variable Y, we can eliminate variables that do not contribute
significantly to the prediction of Y. Backward elimination is an iterative
procedure where the least significant X is “weeded” out until the best
remaining variables are all significant contributors to predicting Y.

X1,
X2,
X1,
X3, Elimin
ation
Elimin Elimin
ation ation
X3
:
Xk
 HOW TO IDENTIFY A
WEAK X
 The weakest X is the variable in an MLR model with the largest P-value
that is also not significant (i.e. p-value< a).

WHAT TO DO WITH A
WEAK X
 Remove X from the set of possible predictors Xi. Recompute Multiple
linear regression model with the remaining set of Xi-1.
 HOW DO YOU RECOGNIZE
THE BEST FIT
 Choose the best fit by looking for the highest adjusted r2 and lowest
standard error.
 All variables in model Bo, B1, B2, etc. all have p-values < a, saying
that such variable’s probability of being randomly related to the
predicted Y is very low (i.e. non-random, something significant
occurrence caused the improbable to happen.)
 EXAMPLE
PROBLEM
Find the best fit regression model for Y for candidate
variables X1, X2, X3, X4, X5 and X6
EXAMPLE
PROBLEM
Find the best fit regression model for Y for candidate
variables X1, X2, X3, X4, X5 and X6

Initial model
X1, X2, X3, X4, X5, X6

X1 out

X2, X3, X4, X5, X6

Next model
EXAMPLE
PROBLEM
Find the best fit regression model for Y for candidate
variables X1, X2, X3, X4, X5 and X6

Current model
X2, X3, X4, X5, X6

X5 out

X2, X3, X4, X6

Next model
EXAMPLE
PROBLEM
Find the best fit regression model for Y for candidate
variables X1, X2, X3, X4, X5 and X6

Current model
X2, X3, X4, X6

X3 out

X2, X4, X6

Next model
EXAMPLE
PROBLEM
Find the best fit regression model for Y for candidate
variables X1, X2, X3, X4, X5 and X6

Current model
X2, X4, X6

X2 out

X4, X6

Next model
EXAMPLE
PROBLEMFind the best fit regression model for Y for candidate
variables X1, X2, X3, X4, X5 and X6

Current model
X4, X6

We have a
winner!
Stopping condition: All variables in model have
significant p-values.
 ALTERNATIVE BEST SUBSETS
CRITERION

 Calculate the value Cp for each potential regression model

 Consider models with Cp values that is less than and closest to

k+1

 k is the number of independent variables in the model under consideration

 ALTERNATIVE BEST SUBSETS
CRITERION
(continued)
 The Cp Statistic

(1  Rk2 )(n  T )
Cp   (n  2(k  1))
1  RT2

Where k = number of independent variables included in a

particular regression model
T = total number of parameters to be estimated in the
full regression model
Rk2 = coefficient of multiple determination for model with k
independent variables
R 2T = coefficient of multiple determination for full model with
all T estimated parameters
Minitab has a “Best Subset” feature in REGRESSION that gives (sort of)
a short-cut way to do variable search.
Minitab has a “Best Subset” feature in REGRESSION that gives (sort of)
a short-cut way to do variable search.
ASPECTS OF MULTIPLE REGRESSION
MODELING
Multicollinearity

Variance Inflation Factor (VIF)

ASPECTS OF MULTIPLE REGRESSION
MODELING
Multicollinearity
The presence of multicollinearity can be detected in several
ways. Two of the more easily understood of these are:

73
Variance
Inflation
Factor (VIF)
EXAMPLE OF (PROBLEMATIC)
MULTICOLLINEARITY

High VIF

If independent Xi variables are

collinear, we only need one
variable to represent any of them.
“Si Mfg o si Export? Mamili ka ng
isa sa amin!”
 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
 Discussed Model Building, and the issue of Multicollinearity
Slides © 2009 Prentice-Hall, Inc.. Basic Business Statistics 11e