Linear Regression

1
Bicycling World, a magazine devoted to ccling, reviews hundreds of biccles throughtout th
Race categor contains reviews of biccles used by riders primarily interested in racing. ONe
important factors in selecting a bicycyle for racing is its weight. THe following data show t
(pounds) and price ($) for ten racing biccles reviewed by the magazine:

Model Weight (lbs) Price ($)

Fierro 7b 17.9 2200
HX 5000 16.2 6350
Durbin Ultralight 15.0 8470
Schmidt 16.0 6300
WSilton Advanced 17.3 4100
bicyclette velo 13.2 8700
Supremo Team 16.3 6100
XTC Racer 17.2 2680
D'Onofrio Pro 17.7 3500
Americans #6 14.2 8100

a. Develop a scatter chart with weight as the independent variable. What does the s
indicate about the relationship between the weight and price of these bicycles?
b. Use the data to develop an estimated regression equation that could be used to e
for a bicycle, given its weight. What is the estimated regression model?
c.
Test whether each of the regression parameters B0 and B1 is equal to zero at a 0.0
signifacance. What are the correct interpretations of the estimated regression pa
these interpretations reasonable?

d. How much of the variation in the prices of the bicycles in the sample does the reg
you estimated in part b explain?
e.
The manufacturers of the D'Onofrio Pro plan to introduce the 15-pound D'Onofri
later this year. Use the regression model you estimated in part a to predict the pr
D'Onofrio Elite
2
In a manufacturing process the assembly line speed (feet per minute) was thought to affe
defective parts found during the inspection process. To test this theory, managers devised
which the same batch of parts was inspected visually at a variety of line speeds. They colle
following data:

Number of Defective
Line Speed (ft/min)
Part Found
20 21
20 19
40 15
30 16
60 14
40 17

a. Develop a scatter chart with line speed aas the independent variable, What does
indicate about the relationship between line speed and the number of defective
b. Use the data to develop an estimated regression equation that could be used to p
number of defective parts found, given the line speed. What is the estimated reg
c.
Test whether each of the regression parameters B0 and B1 is equal to zero at a 0.0
signifacance. What are the correct interpretations of the estimated regression pa
these interpretations reasonable?

d.
How much of the variation in the number of defective parts found for the sample
model you estimated in part b explain?

Multiple Regression
9
Dixie Showtime Movie Theaters, Inc., owns and operates a chain of cinemas in seeral mar
southern United States. The owners would like to estimate weekly gross revenue as a fucn
advertising expenditures. Data for a sample of eight markets for a recent week follows:
 Market Weekly Gross Revenue ($100s) Television Adverti
($100s)
Mobile 101.3
Shreveport 51.9
Jackson 74.8
Birmingham 126.2
Little Rock 137.8
Biloxi 101.4
New Orleans 237.8
Baton Rouge 219.6

a. Develop an estimated regression equation with the amount of television advertis

independent variable. Test for a significant relationship between television adver
gross revenue at the 0.05 level of significance. What is the interpretation of this r
b. How much of the variation in the sample values of weekly gross revenue does the
explain?
c. advertising as the independent variables. Is the overall regression statistically sign
0.05 level of significance? If so, then test whether each of the regression paramet
is equal to zero at a ).05 level of significance. What are the correct interpretation
d. How much of the variation in the sample values of weekly gross revenue does the
explain?
e.
Given the results in parts a and c, what should your next step be? Explain.
f. What are the managerial implications of these results?

10

Resorts & Spas, a magazine devoted to upscale vacations and accommodations, published
Choice LIst of the top 20 independent beachfront boutique hotels in the world. The data s
scores received by these hotels based on the results form Resorts & Spas' annual Readers
Each score represents the percentage of respondents who rated a hotel as excellent or ve
of three criteria (comfort, amenities, and in-house dining). An overall score was also repor
rank the hotels. The highest ranked hotel, the Muri Beach Odyssey, has an overall score o
highest component of which is 97.7 for in-house dining.

Hotel Overall Comfort Amenities

Muri Beach Odyssey 94.3 94.5 90.8
Pattaya Resort 92.9 96.6 84.1
Sojourner’s Respite 92.8 99.9 100.0
Spa Carribe 91.2 88.5 94.7
Penang Resort and Spa 90.4 95.0 87.8
Mokihana Hōkele 90.2 92.4 82.0
Theo’s of Cape Town 90.1 95.9 86.2
Cap d’Agde Resort 89.8 92.5 92.5
Spirit of Mykonos 89.3 94.6 85.8
Turismo del Mar 89.1 90.5 83.2
Hotel Iguana 89.1 90.8 81.9
Sidi Abdel Rahman Palace 89.0 93.0 93.0
Sainte-Maxime Quarters 88.6 92.5 78.2
Rotorua Inn 87.1 93.0 91.6
Club Lapu-Lapu 87.1 90.9 74.9
Terracina Retreat 86.5 94.3 78.0
Hacienda Punta Barco 86.1 95.4 77.3
Rendezvous Kolocep 86.0 94.8 76.4
Cabo de Gata Vista 86.0 92.0 72.2
Sanya Deluxe 85.1 93.4 77.3

a. Determine the estimated multiple linear regression equation that can be used to
overall score given the scores for comfort, amenitie, and in-house dining.
b. Use the F-test to determine the overall significance of the regression relationship
conclusion at the 0.01 level of significance?
c. Use the t test to determine the significance of each independent variable. What i
for each test at the 0.01 level of significance?
d. Remove all independent variables that are not significant at the 0.01 level of sign
estimated regression equation. What is your recommended estimated regression
iccles throughtout the year. Its Road
rested in racing. ONe of the most
ollowing data show the weight
e:

a.
Price ($) VS Weight (lbs)
10000
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0
ble. What does the scatter chart The scatter chart shows a negative relationship between the two vari
e of these bicycles?
at could be used to estimate the price y=-0.0006(x)+19.4912
on model?

equal to zero at a 0.05 level of

mated regression parameters? Are The shows that the intercept has significant use in the regression mo
The data shows that there is a significant relationship between weigh

sample does the regression model The data shows that 86% of the y(prices) are explained by x(weight),

y=-1439.01(x)+28818
e 15-pound D'Onofrio Elite bicycle 7232.9 dollar
rt a to predict the price of the
 a.
was thought to affect the number of Chart Title
25
ry, managers devised a situation in
ne speeds. They collected the
20

15

10

0
15 20 25 30 35 40 45 50 55 60
The scatter chart shows a negative relationship between the

variable, What does the scatter chart

umber of defective parts found?
at could be used to predict the y=-0.14783(x)+22.1739
is the estimated regression model?

equal to zero at a 0.01 level of It shows that the intercept has significant use in the regression mode
mated regression parameters? Are It shows that there is no significant relationship between the line spe

The data shows that 74% of the y(number of defective) are explained
ound for the sample data does the

nemas in seeral markets in the

oss revenue as a fucntion of
ent week follows:
 Television Advertising Newspaper Advertising
($100s) (100s)
5.0 1.5
3.0 3.0
4.0 1.5
4.3 4.3
3.6 4.0
3.5 2.3
5.0 8.4
6.9 5.8

of television advertising as the

een television advertising and weekl
terpretation of this relationship?
oss revenue does the model in part a

ssion statistically significant at the

e regression parameters Bo,B1 and B2
orrect interpretation of the estimated
oss revenue does the model in part c

p be? Explain.

modations, published its Reader's

he world. The data shown are the
Spas' annual Readers' Choice Survey.
tel as excellent or very good on one
score was also reported and used to
as an overall score of 94.3, the

Amenities In-House Dining

 97.7
96.6
88.4
97.0
91.1
98.7
91.9
88.8
90.7
90.4
88.5
89.6
91.2
73.5
89.6
91.5
90.8
91.4
89.2
91.8

that can be used to predict the

ouse dining.
gression relationship. What is the

dent variable. What is the conclusion

he 0.01 level of significane from the

estimated regression equation?
 b. SUMMARY OUTPUT
)
Regression Statistics
Multiple R 0.929366
R Square 0.863722
Adjusted R 0.846687
Standard E 942.2661
Observati 10

ANOVA
df SS MS F Significance F
17.0 18.0 19.0 Regression 1 45017877 45017877 50.7035 9.99374E-05
etween the two variables. Residual 8 7102923 887865
Total 9 52120800

Coefficients
Standard Error t Stat P-value Lower 95% Upper 95%
Intercept 28818 3267.257 8.820244 2.15E-05 21283.6959 36352.31
Weight (lb -1439.01 202.0895 -7.12064 9.99E-05 -1905.02569 -972.987
n the regression model with the level of significance of 0.05
ship between weight and price with a level of significance of 0.05.

lained by x(weight), or 86% of the values fit the model.

 b. SUMMARY OUTPUT

Regression Statistics
Multiple R 0.859727
R Square 0.73913
Adjusted R 0.673913
Standard E 1.489091
Observati 6

ANOVA
df SS MS F Significance F
50 55 60 65 Regression 1 25.13043 25.13043 11.33333333 0.028135
onship between the two variables. Residual 4 8.869565 2.217391
Total 5 34

Coefficients
Standard Error t Stat P-value Lower 95%
Intercept 22.17391 1.652746 13.41641 0.000178521 17.58515
Line Speed -0.14783 0.043911 -3.3665 0.028134748 -0.26974

he regression model with the level of significance of 0.01

between the line speed and number of defective with a level of significance of 0.01.

ective) are explained by x(line speed), or 74% of the values fit the model.
 c. Null Hypothesis B0=0 B0=intercept of the regr
Alternative Hypothesis B0=/0

Level of significance 0.05

T-stat 8.8202443505621
P-value 2.148884268E-05

Conclusion Reject

It shows that the intercept has significant use in the regression model wi

Null Hypothesis B1=0 There is no significant relationsh

Alternative Hypothesis B1=/0 There is a significant relationship
*B1=slope of the regression model
Lower 95.0%
Upper 95.0% Level of significance 0.05
21283.7 36352.31
-1905.03 -972.987 T-stat -7.120638885
P-value 9.993745E-05

Conclusion Reject

It shows that there is a significant relationship between weight and price

 c. Null Hypothesis B0=0
Alternative Hypothesis B0=/0

Level of significance 0.01

T-stat 13.41641
P-value 0.000179

Conclusion Reject

Significance F It shows that the intercept has significant use in the regression m

Null Hypothesis B1=0

Alternative Hypothesis B1=/0
*B1=slope of the regression model
Upper 95%Lower 99.0%
Upper 99.0% Level of significance 0.01
26.76267 14.56451 29.78331
-0.02591 -0.35 0.054344 T-stat -3.3665
P-value 0.0281

Conclusion Do not reject

It shows that there is a significant relationship between the line

 B0=intercept of the regression model

n the regression model with the level of significance of 0.05

s no significant relationship between weight and price.

s a significant relationship between weight and price.

between weight and price with a level of significance of 0.05.

 B0=intercept of the regression model

ant use in the regression model with the level of significance of 0.01.

There is no significant relationship between weight and price.

There is a significant relationship between weight and price.

tionship between the line speed and number of defective with a level of significance of 0.01.