15 ch ken black solution

Chapter 15: Building Multiple Regression Models 1
Chapter 15
Building Multiple Regression Models
LEARNING OBJECTIVES
This chapter presents the potential of multiple regression analysis as a tool in
business decision making and its applications, thereby enabling you to:
1. Analyze and interpret nonlinear variables in multiple regression analysis.
2. Understand the role of qualitative variables and how to use them in
multiple regression analysis.
3. Learn how to build and evaluate multiple regression models.
4. Learn how to detect influencial observations in regression analysis.
CHAPTER TEACHING STRATEGY
In chapter 14, the groundwork was prepared for chapter 15 by presenting
multiple regression models along with mechanisms for testing the strength of the
models such as se, R2
, t tests of the regression coefficients, and the residuals.
The early portion of the chapter is devoted to nonlinear regression models
and search procedures. There are other exotic types of regression models that can
be explored. It is hoped that by studying section 15.1, the student will be
somewhat prepared to explore other nonlinear models on his/her own. Tukey’s
Ladder of Transformations can be useful in steering the research towards
particular recoding schemes that will result in better fits for the data.
Dummy or indicator variables can be useful in multiple regression
analysis. Remember to emphasize that only one dummy variable is used to
represent two categories (yes/no, male/female, union/nonunion, etc.). For c
categories of a qualitative variable, only c-1 indicator variables should be
included in the multiple regression model.

Several search procedures have been discussed in the chapter including
stepwise regression, forward selection, backward elimination, and all possible
regressions. All possible regressions is presented mainly to demonstrate to the
student the large number of possible models that can be examined. Most of the
effort is spent on stepwise regression because of its common usage. Forward
selection is presented as the same as stepwise regression except that forward
selection procedures do not go back and examine variables that have been in the
model at each new step. That is, with forward selection, once a variable is in the
model, it stays in the model. Backward elimination begins with a "full" model of
all predictors. Sometimes there may not be enough observations to justify such a
model.
CHAPTER OUTLINE
15.1 Non Linear Models: Mathematical Transformation
Polynomial Regression
Tukey’s Ladder of Transformations
Regression Models with Interaction
Model Transformation
15.2 Indicator (Dummy) Variables
15.3 Model-Building: Search Procedures
Search Procedures
All Possible Regressions
Stepwise Regression
Forward Selection
Backward Elimination
15.4 Multicollinearity
KEY TERMS
All Possible Regressions Qualitative Variable
Backward Elimination Search Procedures
Dummy Variable Stepwise Regression
Forward Selection Tukey’s Four-quadrant Approach
Indicator Variable Tukey’s Ladder of
Multicollinearity Transformations
Nonlinear Regression Model Variance Inflation Factor

SOLUTIONS TO PROBLEMS IN CHAPTER 15
15.1 Simple Regression Model:
yˆ = - 147.27 + 27.128 x
F = 229.67 with p = .000, se = 27.27, R2
= .97, adjusted R2
= .966, and
t = 15.15 with p = .000. This is a very strong simple regression model.
Quadratic Model (Using both x and x2
):
yˆ = - 22.01 + 3.385 X + 0.9373 x2
F = 578.76 with p = .000, se = 12.3, R2
= .995, adjusted R2
= .993, for x:
t = 0.75 with p = .483, and for x2
: t = 5.33 with p = .002. The quadratic
model is also very strong with an even higher R2
value. However, in this
model only the x2
term is a significant predictor.
15.2 The model is:
yˆ = b0b1
x
Using logs: log y = log b0 + x log b1
The regression model is solved for in the computer using the values of x
and the values of log y. The resulting regression equation is:
log y = 0.5797 + 0.82096 x
F = 68.83 with p = .000, se = 0.1261, R2
= .852, and adjusted R2
= .839.
This model has relatively strong predictability.

15.3 Simple regression model:
Y
ˆˆ = - 1456.6 + 71.017 x
R2
= .928 and adjusted R2
= .910. t = 7.17 with p = .002.
Quadratic regression model:
yˆ = 1012 - 14.06 X + 0.6115 x2
R2
= .947 but adjusted R2
= .911. The t ratio for the x term is t = - 0.17
with p = .876. The t ratio for the x2
term is t = 1.03 with p = .377
Neither predictor is significant in the quadratic model. Also, the adjusted
R2
for this model is virtually identical to the simple regression model. The
quadratic model adds virtually no predictability that the simple regression
model does not already have. The scatter plot of the data follows:
11010090807060504030
7000
6000
5000
4000
3000
2000
1000
Eq & Sup Exp
AdExp

15.4 The model is:
yˆ = b0b1
x
Using logs: log y= log b0 + x log b1
The regression model is solved for in the computer using the values of x
and the values of log y where x is failures and y is liabilities. The resulting
regression equation is:
log liabilities = 3.1256 + 0.012846 failures
F = 19.98 with p = .001, se = 0.2862, R2
= .633.
This model has modest predictability.
15.5 The regression model is:
yˆ = - 28.61 - 2.68 x1 + 18.25 x2 - 0.2135 x1
2
- 1.533 x2
2
+ 1.226 x1*x2
F = 63.43 with p = .000 significant at α = .001
se = 4.669, R2
= .943
None of the t ratios for this model are significant. They are t(x1) = - 0.25
with p = .805, t(x2) = 0.91 with p = .378, t(x1
2
) = - 0.33 with .745,
t(x2
2
) = - 0.68 with .506, and t(x1*x2) = 0.52 with p = .613. This model has
a high R2
yet none of the predictors are individually significant.
The same thing occurs when the interaction term is not in the model.
None of the t tests are significant. The R2
remains high at .957 indicating
that the loss of the interaction term was insignificant.
15.6 The F value shows very strong overall significance with a p-value of
.00000073. This is reinforced by the high R2
of .910 and adjusted R2
of
.878. An examination of the t values reveals that only one of the
regression coefficients is significant at α = 05 and that is the interaction
term with a p-value of .039. Thus, this model with both variables, the
square of both variables, and the interaction term contains only 1
significant t test and that is for interaction.
Without interaction, the R2
drops to .877 and adjusted R2
to .844. With the
interaction term removed, both variable x2 and x2
2
are significant at
α = .01.

15.7 The regression equation is:
yˆ = 13.619 - 0.01201 x1 + 2.998 x2
The overall F = 8.43 is significant at α = .01 (p = .009).
se = 1.245, R2
= .652, adjusted R2
= .575
The t ratio for the x1 variable is only t = -0.14 with p = .893. However the
t ratio for the dummy variable, x2 is t = 3.88 with p = .004. The indicator
variable is the significant predictor in this regression model which has
some predictability (adjusted R2
= .575).
15.8 The indicator variable has c = 4 categories as shown by the c - 1 = 3
categories of the predictors (x2, x3, x4).
The regression equation is:
yˆ = 7.909 + 0.581 x1 + 1.458 x2 - 5.881 x3 - 4.108 x4
Overall F = 13.54, p = .000 significant at α = .001
se = 1.733, R2
= .747
For the predictors, t = 0.56 with p = .585 for the x1 variable (not
significant), t = 1.32 with p = .208 for the first indicator variable (x2) and
is non significant, t = -5.32 with p = .000 for x3 the second indicator
variable and this is significant at α = .001, t = -3.21 with p = .007 for the
third indicator variable (x4) which is significant at α = .01. This model has
strong predictability and the only significant predictor variables are the
two dummy variables, x3 and x4.
15.9 This regression model has relatively strong predictability as indicated by
R2
= .795. Of the three predictor variables, only x1 and x2 have significant
t ratios (using α = .05). x3 (a non indicator variable) is not a significant
predictor. x1, the indicator variable, plays a significant role in this model
along with x2.

yˆ = 41.225 + 1.081 x1 – 18.404 x2
F = 8.23 with p = .0017 which is significant at α = .01. se = 11.744,
R2
= .388 and the adjusted R2
= .341.
The t-ratio for x2 (the dummy variable) is –4.05 which has an associated
p-value of .0004 and is significant at α = .001. The t-ratio of 0.80 for x1
is not significant (p-value = .4316). With x2 = 0, the regression model
becomes yˆ = 41.225 + 1.081x1. With x2 = 1, the regression model
becomes yˆ = 22.821 + 1.081x1. The presence of x2 causes the y
intercept to drop by 18.404. The graph of each of these models (without
the dummy variable and with the dummy variable equal to one) is shown
below:
15.11 The regression equation is:
Price = 7.066 - 0.0855 Hours + 9.614 ProbSeat + 10.507 FQ
The overall F = 6.80 with p = .009 which is significant at α = .01. se =
4.02, R2
= .573. The difference between R2
and
adjusted R2
indicates that there are some non-significant predictors in the
model. The t ratios, t = - 0.56 with p = .587 and t = 1.37 with p = .202, of
Hours and Probability of Being Seated are non-significant at α = .05. The
only significant predictor is the dummy variable, French Quarter or not,
which has a t ratio of 3.97 with p = .003 which is significant at α = .01.
The positive coefficient on this variable indicates that being in the French
Quarter adds to the price of a meal.

15.12 There will be six predictor variables in the regression analysis:
three for occupation, two for industry, and one for marital status. The
independent variable is job satisfaction, which makes a total of seven
variables.
15.13 Stepwise Regression:
Step 1: x2 enters the model, t = - 7.35 and R2
= .794
The model is = 36.15 - 0.146 x2
Step 2: x3 enters the model and x2 remains in the model.
t for x2 is -4.60, t for x3 is 2.93. R2
= .876.
The model is yˆ = 26.40 - 0.101 x2 + 0.116 x3
The variable, x1, never enters the procedure.
15.14 Stepwise Regression:
Step 1: x4 enters the model, t = - 4.20 and R2
= .525
The model is yˆ = 133.53 - 0.78 x4
Step 2: x2 enters the model and x4 remains in the model.
t for x4 is - 3.22 and t for x2 is 2.15. R2
= .637
The model is yˆ = 91.01 - 0.60 x4 + 0.51 x2
The variables, x1 and x3 never enter the procedure.
15.15 The output shows that the final model had four predictor variables, x4, x2,
x5, and x7. The variables, x3 and x6 did not enter the stepwise analysis.
The procedure took four steps. The final model was:
y1 = - 5.00 x4 + 3.22 x2 + 1.78 x5 + 1.56 x7
The R2
for this model was .5929, and se was 3.36. The t ratios were:
tx4 = 3.07, tx2 = 2.05, tx5 = 2.02, and tx7 = 1.98.

15.16 The output indicates that the stepwise process only went two steps.
Variable x3 entered at step one. However, at step two, x3 dropped out of
the analysis and x2 and x4 entered as the predictors. x1 was the dependent
variable. x5 never entered the procedure and was not included in the final
model as x3 was not. The final regression model was:
Yˆ = 22.30 + 12.38 x2 + 0.0047 x4.
R2
= .682 and se = 9.47. tx2 = 2.64 and tx4 = 2.01.
15.17 The output indicates that the procedure went through two steps. At step 1,
dividends entered the process yielding an R2
of .833 by itself. The t value
was 6.69 and the model was yˆ = - 11.062 + 61.1 x1. At step 2, net
income entered the procedure and dividends remained in the model. The
R2
for this two predictor model was .897 which is a modest increase from
the simple regression model shown in step one. The step 2 model was:
Premiums earned = - 3.726 + 45.2 dividends + 3.6 net income
For step 2, tdividends = 4.36 (p-value = .002)
and tnet income = 2.24 (p-value = .056).
15.18 This stepwise regression procedure only went one step. The only
significant predictor was natural gas. No other predictors entered the
model. The regression model is:
Electricity = 1.748 + 0.994 Natural Gas
For this model, R2
= .9295 and se = 0.490. The t value for natural gas was
11.48.

15.19 y x1 x2 x3
y - -.653 -.891 .821
x1 -.653 - .650 -.615
x2 -.891 .650 - -.688
x3 .821 -.615 -.688 -
There appears to be some correlation between all pairs of the predictor
variables, x1, x2, and x3. All pairwise correlations between independent
variables are in the .600 to .700 range.
15.20 y x1 x2 x3 x4
y - -.241 .621 .278 -.724
x1 -.241 - -.359 -.161 .325
x2 .621 -.359 - .243 -.442
x3 .278 -.161 .243 - -.278
x4 -.724 .325 -.442 -.278 -
An examination of the intercorrelations of the predictor variables reveals
that the highest pairwise correlation exists between variables x2 and
x4 (-.442). Other correlations between independent variables are less than
.400. Multicollinearity may not be a serious problem in this regression
analysis.

15.21 The stepwise regression analysis of problem 15.17 resulted in two of the
three predictor variables being included in the model. The simple
regression model yielded an R2
of .833 jumping to .897 with the two
predictors. The predictor intercorrelations are:
Net
Income Dividends Gain/Loss
Net - .682 .092
Income
Dividends .682 - -.522
Gain/Loss .092 -.522 -
An examination of the predictor intercorrelations reveals that Gain/Loss
and Net Income have very little correlation, but Net Income and Dividends
have a correlation of .682 and Dividends and Gain/Loss have a correlation
of -.522. These correlations might suggest multicollinearity.
15.22 The intercorrelations of the predictor variables are:
Natural Fuel
Gas Oil Gasoline
Natural
Gas - .570 .701
Fuel Oil .570 - .934
Gasoline .701 .934 -
Each of these intercorrelations is not small. Of particular concern is the
correlation between fuel oil and gasoline which is .934. These two
variables seem to be adding about the same predictability to the model. In
the stepwise regression analysis only natural gas entered the procedure.
Perhaps the overlapping information between natural gas and fuel oil and
gasoline was such that fuel oil and gasoline did not have significant unique
variance to add to the prediction.

yˆ = 564 - 27.99 x1 - 6.155 x2 - 15.90 x3
F = 11.32 with p = .003, se = 42.88, R2
= .809, adjusted R2
= .738. For x1,
t = -0.92 with p = .384, for x2, t = -4.34 with p = .002, for x3, t = -0.71 with
p = .497. Thus, only one of the three predictors, x2, is a significant
predictor in this model. This model has very good predictability (R2
=
.809). The gap between R2
and adjusted R2
underscores the fact that there
are two nonsignificant predictors in this model. x1 is a nonsignificant
indicator variable.
15.24 The stepwise regression process included two steps. At step 1, x1 entered
the procedure producing the model:
yˆ = 1540 + 48.2 x1.
The R2
at this step is .9112 and the t ratio is 11.55. At step 2, x1
2
entered
the procedure and x1 remained in the analysis. The stepwise regression
procedure stopped at this step and did not proceed. The final model was:
yˆ = 1237 + 136.1 x1 - 5.9 x1
2
.
The R2
at this step was .9723, the t ratio for x1 was 7.89, and the t ratio for
x1
2
was - 5.14.
15.25 In this model with x1 and the log of x1 as predictors, only the log x1 was a
significant predictor of y. The stepwise procedure only went to step 1.
The regression model was:
yˆ = - 13.20 + 11.64 Log x1. R2
= .9617 and the t ratio of Log X1 was
17.36. This model has very strong predictability using only the log of
the X1 variable.

Grain = - 4.675 + 0.4732 Oilseed + 1.18 Livestock
The value of R2
was .901 and adjusted R2
= .877.
se = 1.761. F = 36.55 with p = .000.
toilseed = 3.74 with p = .006 and tlivestock = 3.78 with p = .005. Both
predictors are significant at α = .01. This is a model with strong
predictability.
15.27 The stepwise regression procedure only used two steps. At step 1, Silver
was the lone predictor. The value of R2
was .5244. At step 2, Aluminum
entered the model and Silver remained in the model. However, the R2
jumped to .8204. The final model at step 2 was:
Gold = - 50.19 + 18.9 Silver +3.59 Aluminum.
The t values were: tSilver = 5.43 and tAluminum = 3.85.
Copper did not enter into the process at all.
15.28 The regression model was:
Employment = 71.03 + 0.4620 NavalVessels + 0.02082 Commercial
F = 1.22 with p = .386 (not significant)
R2
= .379 and adjusted R2
= .068
The low value of adjusted R2
indicates that the model has very low
predictability. Both t values are not significant (tNavalVessels = 0.67 with
p = .541 and tCommercial = 1.07 with p = .345). Neither predictor is a
significant predictor of employment.

15.29 There were four predictor variables. The stepwise regression procedure
went three steps. The predictor, apparel, never entered in the stepwise
process. At step 1, food entered the procedure producing a model with an
R2
of .84. At step 2, fuel oil entered and food remained. The R2
increased
to .95. At step 3, shelter entered the procedure and both fuel oil and food
remained in the model. The R2
at this step was .96. The final model was:
All = -1.0615 + 0.474 Food + 0.269 Fuel Oil + 0.249 Shelter
The t ratios were: tfood = 8.32, tfuel oil = 2.81, tshelter = 2.56.
15.30 The stepwise regression process with these two independent variables only
went one step. At step 1, Soybeans entered in producing the model,
Corn = - 2,962 + 5.4 Soybeans. The R2
for this model was .7868.
The t ratio for Soybeans was 5.43. Wheat did not enter in to the analysis.
15.31 The regression model was:
Grocery = 76.23 + 0.08592 Housing + 0.16767 Utility
+ 0.0284 Transportation - 0.0659 Healthcare
F = 2.29 with p = .095 which is not significant at α = .05.
se = 4.416, R2
= .177.
Only one of the four predictors has a significant t ratio and that is Utility
with t = 2.57 and p = .018. The ratios and their respective probabilities
are:
thousing = 1.68 with p = .109, ttransportation = 0.17 with p = .87, and
thealthcare = - 0.64 with p = .53.
This model is very weak. Only the predictor, Utility, shows much promise
in accounting for the grocery variability.

15.32 The output suggests that the procedure only went two steps.
At step 1, x1 entered the model yielding an R2
of .7539. At step 2,
x2 entered the model and x1 remained. The procedure stopped here with a
final model of:
yˆ = 124.5 - 43.4 x1 + 1.36 x2
The R2
for this model was .8059 which indicates relatively strong
predictability with two independent variables. Since there were four
predictor variables, two of the variables did not enter the stepwise process.
15.33 Of the three predictors, x2 is an indicator variable. An examination of the
stepwise regression output reveals that there were three steps and that all
three predictors end up in the final model. Variable x3 is the strongest
individual predictor of y and entered at step one resulting in an R2
of
.8124. At step 2, x2 entered the process and variable x3 remained in the
model. The R2
at this step was .8782. At step 3, variable x1 entered the
procedure. Variables x3 and x2 remained in the model. The final R2
was
.9407. The final model was:
yˆ = 87.89 + 0.071 x3 - 2.71 x2 - 0.256 x1
15.34 The R2
for the full model is .321. After dropping out variable, x3, the R2
is
still .321. Variable x3 added virtually no information to the model. This is
underscored by the fact that the p-value for the t test of the slope for x3 is
.878 indicating that there is no significance. The standard error of the
estimate actually drops slightly after x3 is removed from the model.

15 ch ken black solution

More Related Content

What's hot (20)

Similar to 15 ch ken black solution (20)

Recently uploaded (20)

15 ch ken black solution