Anda di halaman 1dari 50

Chapter14

SimpleLinearRegression

LearningObjectives

1. Understandhowregressionanalysiscanbeusedtodevelopanequationthatestimates
mathematicallyhowtwovariablesarerelated.

2. Understandthedifferencesbetweentheregressionmodel,theregressionequation,andthe
estimatedregressionequation.

3. Knowhowtofitanestimatedregressionequationtoasetofsampledatabasedupontheleast
squaresmethod.

4. Beabletodeterminehowgoodafitisprovidedbytheestimatedregressionequationandcompute
thesamplecorrelationcoefficientfromtheregressionanalysisoutput.

5. Understandtheassumptionsnecessaryforstatisticalinferenceandbeabletotestforasignificant
relationship.

6. Knowhowtodevelopconfidenceintervalestimatesofygivenaspecificvalueofxinboththecase
ofameanvalueofyandanindividualvalueofy.

7. Learnhowtousearesidualplottomakeajudgementastothevalidityoftheregression
assumptions.

8. Knowthedefinitionofthefollowingterms:

independentanddependentvariable
simplelinearregression
regressionmodel
regressionequationandestimatedregressionequation
scatterdiagram
coefficientofdetermination
standarderroroftheestimate
confidenceinterval
predictioninterval
residualplot

141
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

Solutions:

1 a.

16
14
12
10
8
y

6
4
2
0
0 1 2 3 4 5 6
x

b. Thereappearstobeapositivelinearrelationshipbetweenxandy.

c. Manydifferentstraightlinescanbedrawntoprovidealinearapproximationofthe
relationshipbetweenxandy;inpart(d)wewilldeterminetheequationofastraightline
thatbestrepresentstherelationshipaccordingtotheleastsquarescriterion.

Sxi 15 Syi 40
d. x= = =3 y= = =8
n 5 n 5

S( xi - x )( yi - y ) = 26 S( xi - x ) 2 = 10

S( xi - x )( yi - y ) 26
b1 = = = 2.6
S( xi - x ) 2 10

b0 = y - b1 x = 8 - (2.6)(3) = 0.2

y= 0.2 2.6 x

e. y= 0.2 2.6(4) = 10.6

142
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

2. a.

b. Thereappearstobeanegativelinearrelationshipbetweenxandy.

c. Manydifferentstraightlinescanbedrawntoprovidealinearapproximationofthe
relationshipbetweenxandy;inpart(d)wewilldeterminetheequationofastraightline
thatbestrepresentstherelationshipaccordingtotheleastsquarescriterion.

Sxi 55 Syi 175


d. x= = = 11 y= = = 35
n 5 n 5

S( xi - x )( yi - y ) = -540 S( xi - x ) 2 = 180

S( xi - x )( yi - y ) -540
b1 = = = -3
S( xi - x ) 2 180

b0 = y - b1 x = 35 - (-3)(11) = 68

y = 68 - 3 x

e. y = 68 - 3(10) = 38

143
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

3. a.

Sxi 50 Syi 83
b. x= = = 10 y= = = 16.6
n 5 n 5

S( xi - x )( yi - y ) = 171 S( xi - x ) 2 = 190

S( xi - x )( yi - y ) 171
b1 = = = 0.9
S( xi - x ) 2 190

b0 = y - b1 x = 16.6 - (0.9)(10) = 7.6

y = 7.6 0.9 x

c. y = 7.6 0.9(6) = 13

144
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

4. a.

135
130
Weight 125

120
115
110
105
100
61 62 63 64 65 66 67 68 69
Height

b. Thereappearstobeapositivelinearrelationshipbetweenx=heightandy=weight.

c. Manydifferentstraightlinescanbedrawntoprovidealinearapproximationofthe
relationshipbetweenxandy;inpart(d)wewilldeterminetheequationofastraightline
thatbestrepresentstherelationshipaccordingtotheleastsquarescriterion.

Sxi 325 Syi 585


d. x= = = 65 y= = = 117
n 5 n 5

S( xi - x )( yi - y ) = 110 S( xi - x ) 2 = 20

S( xi - x )( yi - y ) 110
b1 = = = 5.5
S( xi - x ) 2 20

b0 = y - b1 x = 117 - (5.5)(65) = -240.5

y = -240.5 55
. x

e. y = -240.5 55
. x = -240.5 55
. (63) = 106 pounds

145
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

5. a.

b. Thereappearstobeapositiverelationshipbetweenpriceandrating.ThesignthatsaysQuality:
YouGetWhatYouPayFordoesfairlyreflectthepricequalityrelationshipforellipticals.

c. Letx=price($)andy=rating.

Sxi 1500 Syi 592


x= = = 1875 y= = = 74
n 8 n 8

S( xi - x )( yi - y ) = 68,900 S( xi - x )2 = 8,155,000

S( xi - x )( yi - y ) 68,900
b1 = = = .008449
S( xi - x ) 2 8,155,000

b0 = y - b1 x = 74 - (.008449)(1875) = 58.158

y = 58.158 .008449 x

d. y = 58.158 .008449 x = 58.158 .008449(1500) = 70.83 orapproximately71

146
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

6. a.

b. There appears to be a negative linear relationship between x = miles and y = sales price.

If the car has higher miles, the sales price tends to be lower.

Sxi 874 Syi 66.4


c. x= = = 87.4 y= = = 6.64
n 10 n 10

S( xi - x )( yi - y ) = -135.66 S( xi - x )2 = 5152.4

S( xi - x )( yi - y ) -135.66
b1 = = = -.02633
S( xi - x ) 2 5152.40

b0 = y - b1 x = 6.64 - (-.02633)(87.4) = 8.9412

y = 8.9412 - .02633 x

d. The slope of the estimated regression equation is -.02633. Thus, a one unit increase in the value of
x will result in a decrease in the estimated value of y equal to .02633. Because the data were
recorded in thousands, every additional 1000 miles on the cars odometer will result in a $26.33
decrease in the estimated price.

e. y = 8.9412 - .02633(100) = 6.3 or $6300

147
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

7. a.

150
140
Annual Sales ($1000s) 130
120
110
100
90
80
70
60
50
0 2 4 6 8 10 12 14
Years of Experience

b. Letx=yearsofexperienceandy=annualsales($1000s)

Sxi 70 Syi 1080


x= = =7 y= = = 108
n 10 n 10

S( xi - x )( yi - y ) = 568 S( xi - x ) 2 = 142

S( xi - x )( yi - y ) 568
b1 = = =4
S( xi - x ) 2 142

b0 = y - b1 x = 108 - (4)(7) = 80

y = 80 4 x

c. y = 80 4 x = 80 4(9) = 116 or$116,000

148
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

8. a.

b. The scatter diagram and the slope of the estimated regression equation indicate a negative linear
relationship between x = temperature rating and y = price. Thus, it appears that sleeping bags with
a lower temperature rating cost more than sleeping bags with a higher temperature rating. In other
words, it costs more to stay warmer.

c. x = Sxi / n = 209 /11 = 19 y = Syi / n = 2849 /11 = 259

S( xi - x )( yi - y ) = -10,090 S( xi - x ) 2 = 1912

S( xi - x )( yi - y ) -10, 090
b1 = = = -5.2772
S( xi - x ) 2 1912

b0 = y - b1 x = 259 - (-5.2772)(19) = 359.2668

y = 359.2668 - 5.2772 x

d. y = 359.2668 - 5.2772 x = 359.2668 - 5.2772(20) = 253.72

Thus, the estimate of the price of sleeping bag with a temperature rating of 20 is approximately
$254.

149
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

9. a.

b. There appears to be a positive linear relationship between x = price and y = score.

Sxi 2638 Syi 672


c. x= = = 263.8 y= = = 67.2
n 10 n 10

S( xi - x )( yi - y ) = 14,601.40 S( xi - x ) 2 = 258,695.60

S( xi - x )( yi - y ) 14,601.40
b1 = = = .05644
S( xi - x )2 258,695.60

b0 = y - b1 x = 67.2 - (.05644)(263.8) = 52.311

y = 52.311 .05644 x

d. The slope is .05644. For a $100 higher price, the score can be expected to increase 100(.05644) =
5.644, or about 6 points.

e. y = 52.311 .05644(225) = 65

1410
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

10. a.

b. There appears to be a positive linear relationship between x = age and y = salary.

Sxi 885 Syi 30,939


c. x= = = 59 y= = = 2062.6
n 15 n 15

S( xi - x )( yi - y ) = 175, 265 S( xi - x ) 2 = 1174

S( xi - x )( yi - y ) 175, 265
b1 = = = 149.2888
S( xi - x ) 2 1174

b0 = y - b1 x = 2062.6 - (149.2888)(59) = -6745.44

y = -6745.44 149.29 x

d. y = -6745.44 149.29 x = -6745.44 149.29(72) = 4003 or $4,003,000

11. a.

1411
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

b. There appears to be a positive linear relationship between x = price and y = road-test score.

Sxi 339.6 Syi 930


c. x= = = 28.3 y= = = 77.5
n 12 n 12

S( xi - x )( yi - y ) = 309.90 S( xi - x ) 2 = 346.38

S( xi - x )( yi - y ) 309.90
b1 = = = .8947
S( xi - x ) 2 346.38

b0 = y - b1 x = 77.5 - (.8947)(28.3) = 52.18

y = 52.18 .8947 x

d. The slope is .8947. A sporty car that has a ten thousand dollar higher price can be expected to
have a 10(.8947) = 8.947, or approximately a 9 point higher road-test score.

e. y = 52.18 .8947(36.7) = 85

1412
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

12. a.

b. The scatter diagram indicates a positive linear relationship between x = weight and y = price. Thus,
it appears that PWCs that weigh more have a higher price.

c. x = Sxi / n = 7730 /10 = 773 y = Syi / n = 92, 200 /10 = 9220

S( xi - x )( yi - y ) = 332, 400 S( xi - x )2 = 14,810

S( xi - x )( yi - y ) 332, 400
b1 = = = 22.4443
S( xi - x ) 2 14,810

b0 = y - b1 x = 9220 - (22.4443)(773) = -8129.4439

y = -8129.4439 22.4443x

d. y = -8129.4439 22.4443 x = -8129.4439 22.4443(750) = 8703.78

Thus, the estimate of the price of Jet Ski with a weight of 750 pounds is approximately $8704.

e. No. The relationship between weight and price is not deterministic.

f. The weight of the Kawasaki SX-R 800 is so far below the lowest weight for the data used to
develop the estimated regression equation that we would not recommend using the estimated
regression equation to predict the price for this model.

13. a.

1413
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

30.0
Reasonable Amount of Itemized
Deductions ($1000s) 25.0

20.0

15.0

10.0

5.0

0.0
0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0
Adjusted Gross Income ($1000s)

b. Letx=adjustedgrossincomeandy=reasonableamountofitemizeddeductions

Sxi 399 Syi 97.1


x = = = 57 y= = = 13.8714
n 7 n 7

S( xi - x )( yi - y ) = 1233.7 S( xi - x ) 2 = 7648

S( xi - x )( yi - y ) 1233.7
b1 = = = 0.1613
S( xi - x ) 2 7648

b0 = y - b1 x = 13.8714 - (0.1613)(57) = 4.6773

y = 4.68 016
. x

c. y = 4.68 016
. x = 4.68 016
. (52.5) = 13.08 orapproximately$13,080.

Theagent'srequestforanauditappearstobejustified.

1414
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

14. a.

b. There appears to be a positive linear relationship between x = features rating and y = PCW World
Rating.

Sxi 784 Syi 777


c. x= = = 78.4 y= = = 77.7
n 10 n 10

S( xi - x )( yi - y ) = 147.20 S( xi - x )2 = 284.40

S( xi - x )( yi - y ) 147.20
b1 = = = .51758
S( xi - x ) 2 284.40

b0 = y - b1 x = 77.7 - (.51758)(78.4) = 37.1217

y = 37.1217 .51758 x

d. y = 37.1217 .51758(70) = 73.35 or 73

15. a. Theestimatedregressionequationandthemeanforthedependentvariableare:

yi = 0.2 2.6 xi y =8

Thesumofsquaresduetoerrorandthetotalsumofsquaresare

SSE = ( yi - yi ) 2 = 12.40 SST = ( yi - y ) 2 = 80

Thus,SSR=SSTSSE=8012.4=67.6

b. r2=SSR/SST=67.6/80=.845

Theleastsquareslineprovidedaverygoodfit;84.5%ofthevariabilityinyhasbeenexplainedby
theleastsquaresline.

1415
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

c. rxy = .845 = .9192

16. a. Theestimatedregressionequationandthemeanforthedependentvariableare:

yi = 68 - 3 x y = 35

Thesumofsquaresduetoerrorandthetotalsumofsquaresare

SSE = ( yi - yi ) 2 = 230 SST = ( yi - y ) 2 = 1850

Thus,SSR=SSTSSE=1850230=1620

b. r2=SSR/SST=1620/1850=.876

Theleastsquareslineprovidedanexcellentfit;87.6%ofthevariabilityinyhasbeenexplainedby
theestimatedregressionequation.

c. rxy = .876 = -.936

Note:thesignforrisnegativebecausetheslopeoftheestimatedregressionequationisnegative.
(b1=3)

17. Theestimatedregressionequationandthemeanforthedependentvariableare:

yi = 7.6 .9 x y = 16.6

Thesumofsquaresduetoerrorandthetotalsumofsquaresare

SSE = ( yi - yi ) 2 = 127.3 SST = ( yi - y ) 2 = 281.2

Thus,SSR=SSTSSE=281.2127.3=153.9

r2=SSR/SST=153.9/281.2=.547

Weseethat54.7%ofthevariabilityinyhasbeenexplainedbytheleastsquaresline.

rxy = .547 = .740

18. a. Theestimatedregressionequationandthemeanforthedependentvariableare:

y = 1790.5 581.1x y = 3650

Thesumofsquaresduetoerrorandthetotalsumofsquaresare

SSE = ( yi - y
i ) = 85,135.14
2
SST = ( yi - y ) 2 = 335,000

Thus,SSR=SSTSSE=335,00085,135.14=249,864.86

b. r2=SSR/SST=249,864.86/335,000=.746

1416
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

Weseethat74.6%ofthevariabilityinyhasbeenexplainedbytheleastsquaresline.
c. rxy = .746 = .8637

19. a. Theestimatedregressionequationandthemeanforthedependentvariableare:

y = 80 + 4x y = 108

Thesumofsquaresduetoerrorandthetotalsumofsquaresare

SSE = ( yi - y i ) 2 = 170 SST = ( yi - y ) 2 = 2442

Thus,SSR=SSTSSE=2442170=2272

b. r2=SSR/SST=2272/2442=.93

Weseethat93%ofthevariabilityinyhasbeenexplainedbytheleastsquaresline.

c. rxy = .93 = .96

20. a.

The scatter diagram indicates a positive linear relationship between x = price and y = score.

x = Sxi / n = 29, 600 /10 = 2960 y = Syi / n = 496 /10 = 49.6

S( xi - x )( yi - y ) = 34,840 S( xi - x ) 2 = 2, 744, 000

S( xi - x )( yi - y ) 34,840
b1 = = = .012697
S( xi - x ) 2
2, 744,000

b0 = y - b1 x = 49.6 - (.012697)(2960) = 12.0169

1417
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

y = 12.0169 .0127 x
b. Thesumofsquaresduetoerrorandthetotalsumofsquaresare

SSE = ( yi - yi ) 2 = 540.0446 SST = ( yi - y ) 2 = 982.4

Thus,SSR=SSTSSE=982.4540.0446=442.3554

r2=SSR/SST=442.3554/982.4=.4503

Thefitprovidedbytheestimatedregressionequationisnotthatgood;only45.03%ofthe
variabilityinyhasbeenexplainedbytheleastsquaresline.

c. y = 12.0169 .0127 x = 12.0169 .0127(3200) = 52.66

The estimate of the overall score for a 42-inch plasma television is approximately 53.

Sxi 3450 Syi 33, 700


21. a. x= = = 575 y= = = 5616.67
n 6 n 6

S( xi - x )( yi - y ) = 712,500 S( xi - x ) 2 = 93, 750

S( xi - x )( yi - y ) 712,500
b1 = = = 7.6
S( xi - x ) 2 93, 750

b0 = y - b1 x = 5616.67 - (7.6)(575) = 1246.67

y = 1246.67 7.6 x

b. $7.60

c. Thesumofsquaresduetoerrorandthetotalsumofsquaresare:

SSE = ( yi - y
i ) = 233,333.33
2
SST = ( yi - y ) 2 = 5, 648,333.33

Thus,SSR=SSTSSE=5,648,333.33233,333.33=5,415,000

r2=SSR/SST=5,415,000/5,648,333.33=.9587

Weseethat95.87%ofthevariabilityinyhasbeenexplainedbytheestimatedregressionequation.

d. y = 1246.67 7.6 x = 1246.67 7.6(500) = $5046.67

22. a. y = 74 SSE = 173.88

Thetotalsumofsquaresis

SST = ( yi - y ) 2 = 756

1418
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

Thus,SSR=SSTSSE=756173.88=582.12

r2=SSR/SST=582.12/756=.77

b. Theestimatedregressionequationprovidedagoodfitbecause77%ofthevariabilityinyhasbeen
explainedbytheleastsquaresline.

c. rxy = .77 = .88

This reflects a strong positive linear relationship between price and rating.

23. a. s2=MSE=SSE/(n2)=12.4/3=4.133

b. s = MSE = 4.133 = 2.033

c. S( xi - x ) 2 = 10

s 2.033
sb1 = = = 0.643
S( xi - x ) 2
10

b1 2.6
d. t= = = 4.044
sb1 .643

Usingttable(3degreesoffreedom),areaintailisbetween.01and.025

pvalueisbetween.02and.05

UsingExcelorMinitab,thepvaluecorrespondingtot=4.04is.0272.

Becausepvalue ,werejectH0:1=0

e. MSR=SSR/1=67.6

F=MSR/MSE=67.6/4.133=16.36

UsingFtable(1degreeoffreedomnumeratorand3denominator),pvalueisbetween.025and.05

UsingExcelorMinitab,thepvaluecorrespondingtoF=16.36is.0272.

Becausepvalue ,werejectH0:1=0

Source Sum Degrees Mean


ofVariation ofSquares ofFreedom Square F pvalue
Regression 67.6 1 67.6 16.36 .0272
Error 12.4 3 4.133
Total 80.0 4

24. a. s2=MSE=SSE/(n2)=230/3=76.6667

b. s = MSE = 76.6667 = 8.7560

1419
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

c. S( xi - x ) 2 = 180

s 8.7560
sb1 = = = 0.6526
S( xi - x ) 2
180

b1 -3
d. t= = = -4.59
sb1 .653

Usingttable(3degreesoffreedom),areaintailislessthan.01;pvalueislessthan.02

UsingExcelorMinitab,thepvaluecorrespondingtot=4.59is.0193.

Becausepvalue ,werejectH0:1=0

e. MSR=SSR/1=1620

F=MSR/MSE=1620/76.6667=21.13

UsingFtable(1degreeoffreedomnumeratorand3denominator),pvalueislessthan.025

UsingExcelorMinitab,thepvaluecorrespondingtoF=21.13is.0193.

Becausepvalue ,werejectH0:1=0

Source Sum Degrees Mean


ofVariation ofSquares ofFreedom Square F pvalue
Regression 230 1 230 21.13 .0193
Error 1620 3 76.6667
Total 1850 4

25. a. s2=MSE=SSE/(n2)=127.3/3=42.4333

s = MSE = 42.4333 = 6.5141

b. S( xi - x ) 2 = 190

s 6.5141
sb1 = = = 0.4726
S( xi - x ) 2
190

b1 .9
t= = = 1.90
sb1 .4726

Usingttable(3degreesoffreedom),areaintailisbetween.05and.10

pvalueisbetween.10and.20

UsingExcelorMinitab,thepvaluecorrespondingtot=1.90is.1530.

1420
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

Becausepvalue> ,wecannotrejectH0:1=0;xandydonotappeartoberelated.

c. MSR=SSR/1=153.9/1=153.9

F=MSR/MSE=153.9/42.4333=3.63

UsingFtable(1degreeoffreedomnumeratorand3denominator),pvalueisgreaterthan.10

UsingExcelorMinitab,thepvaluecorrespondingtoF=3.63is.1530.

Becausepvalue> ,wecannotrejectH0:1=0;xandydonotappeartoberelated.

26. a. Insolvingexercise18,wefoundSSE=85,135.14

s2=MSE=SSE/(n2)=85,135.14/4=21,283.79

s = MSE = 21,283.79 = 14589


.

S( xi - x ) 2 = 0.74

s 145.89
sb1 = = = 169.59
S( xi - x ) 2
0.74

b1 581.1
t= = = 3.43
sb1 169.59

Usingttable(4degreesoffreedom),areaintailisbetween.01and.025

pvalueisbetween.02and.05

UsingExcelorMinitab,thepvaluecorrespondingtot=3.43is.0266.

Becausepvalue ,werejectH0:1=0;thereisasignificantrelationshipbetweengradepoint
averageandmonthlysalary

b. MSR=SSR/1=249,864.86/1=249,864.86

F=MSR/MSE=249,864.86/21,283.79=11.74

UsingFtable(1degreeoffreedomnumeratorand4denominator),pvalueisbetween.025and.05

UsingExcelorMinitab,thepvaluecorrespondingtoF=11.74is.0266.

Becausepvalue ,werejectH0:1=0

c.
Source Sum Degrees Mean
ofVariation ofSquares ofFreedom Square F pvalue
Regression 249864.86 1 249864.86 11.74 .0266
Error 85135.14 4 21283.79

1421
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

Total 335000 5

Sxi 37 Syi 1654


27. a. x= = = 3.7 y= = = 165.4
n 10 n 10

S( xi - x )( yi - y ) = 315.2 S( xi - x ) 2 = 10.1
S( xi - x )( yi - y ) 315.2
b1 = = = 31.2079
S( xi - x ) 2 10.1

b0 = y - b1 x = 165.4 - (31.2079)(3.7) = 49.9308

y = 49.9308 31.2079 x

b. SSE= S( yi - y
i ) = 2487.66 SST= S ( yi - y ) =12,324.4
2 2

Thus,SSR=SSTSSE=12,324.42487.66=9836.74

MSR=SSR/1=9836.74

MSE=SSE/(n2)=2487.66/8=310.96

F=MSR/MSE=9836.74/310.96=31.63

UsingFtable(1degreeoffreedomnumeratorand8denominator),pvalueislessthan.01

UsingExcelorMinitab,thepvaluecorrespondingtoF=31.63is.001.

Becausepvalue ,werejectH0:1=0

Uppersupportandpricearerelated.

c. r2=SSR/SST=9,836.74/12,324.4=.80

Theestimatedregressionequationprovidedagoodfit;weshouldfeelcomfortableusingthe
estimatedregressionequationtoestimatethepricegiventheuppersupportrating.

d. y=49.93+31.21(4)=174.77

28. The sum of squares due to error and the total sum of squares are

SSE = ( yi - yi ) 2 = 12,953.09 SST = ( yi - y ) 2 = 66, 200

Thus, SSR = SST - SSE = 66,200 12,953.09 = 53,246.91

s2 = MSE = SSE / (n - 2) = 12,953.09 / 9 = 1439.2322

s = MSE = 1439.2322 = 37.9372

We can use either the t test or F test to determine whether temperature rating and price are related.

We will first illustrate the use of the t test.

1422
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

Note: from the solution to exercise 10 S( xi - x ) 2 = 1912


s 37.9372
sb1 = = = .8676
S( xi - x ) 2
1912

b1 -5.2772
t= = = -6.0825
sb 1
.8676

Using t table (9 degrees of freedom), area in tail is less than .005; p-value is less than .01

Using Excel or Minitab, the p-value corresponding to t = -6.0825 is .000.

Because p-value , we reject H0: 1 = 0

Because we can reject H0: 1 = 0 we conclude that temperature rating and price are related.

Next we illustrate the use of the F test.

MSR = SSR / 1 = 53,246.91

F = MSR / MSE = 53,246.91 / 1439.2322 = 37.00

Using F table (1 degree of freedom numerator and 9 denominator), p-value is less than .01

Using Excel or Minitab, the p-value corresponding to F = 37.00 is .000.

Because p-value , we reject H0: 1 = 0

Because we can reject H0: 1 = 0 we conclude that temperature rating and price are related.

The ANOVA table is shown below.

Source Sum Degrees Mean


ofVariation ofSquares ofFreedom Square F pvalue
Regression 53,246.91 1 53,246.91 37.00 .000
Error 12,953.09 9 1439.2322
Total 66,200 10

29. SSE= S( yi - yi )2 = 233,333.33SST= S( yi - y ) 2 =5,648,333.33

Thus,SSR=SSTSSE=5,648,333.33233,333.33=5,415,000

MSE=SSE/(n2)=233,333.33/(62)=58,333.33

MSR=SSR/1=5,415,000

F=MSR/MSE=5,415,000/58,333.25=92.83

Sourceof Sum Degreesof Mean


Variation ofSquares Freedom Square F pvalue
Regression 5,415,000.00 1 5,415,000 92.83 .0006
Error 233,333.33 4 58,333.33

1423
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

Total 5,648,333.33 5

UsingFtable(1degreeoffreedomnumeratorand4denominator),pvalueislessthan.01

UsingExcelorMinitab,thepvaluecorrespondingtoF=92.83is.0006.

Becausepvalue ,werejectH0:1=0.Productionvolumeandtotalcostarerelated.

30. SSE= S( yi - yi )2 = 173.88SST= S( yi - y ) 2 =756

Thus,SSR=SSTSSE=756173.88=582.12

s2=MSE=SSE/(n2)=173.88/6=28.98

s = 28.98 = 5.3833

( xi - x ) 2 =8,155,000

s 5.3833
sb1 = = = .001885
S( xi - x ) 2
8,155,000

b1 .008449
t= = = 4.48
sb1 .001885

Usingttable(1degreeoffreedomnumeratorand8denominator),areaintailislessthan.005

pvalueislessthan.01

UsingExcelorMinitab,thepvaluecorrespondingtot=4.48is.0042.

Becausepvalue ,werejectH0:1=0

Thereisasignificantrelationshipbetweenpriceandrating.

31. SSE = 540.04 and SST = 982.40

Thus, SSR = SST - SSE = 982.40 540.04 = 442.36

s2 = MSE = SSE / (n - 2) = 540.04 / 8 = 67.5050

s = MSE = 67.5050 = 8.216

MSR = SSR / 1 = 442.36

F = MSR / MSE = 442.36 / 67.5050 = 6.55

Using F table (1 degree of freedom numerator and 8 denominator), p-value is between .025 and .05

Using Excel or Minitab, the p-value corresponding to F = 6.55 is .034.

Because p-value , we reject H0: 1 = 0

1424
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

Conclusion: price and overall score are related

32. a. s=2.033

x =3 S( xi - x ) 2 = 10

1 ( xp - x )
2
1 (4 - 3) 2
s yp = s = 2.033 = 1.11
n S( xi - x ) 2
5 10

b. y = 0.2 2.6 x = 0.2 2.6(4) = 10.6

y p t / 2 s y p

10.63.182(1.11)=10.63.53

or7.07to14.13

1 ( xp - x )
2
1 (4 - 3) 2
c. sind = s 1 = 2.033 1 = 2.32
n S( xi - x ) 2 5 10

d. y p t / 2 sind

10.63.182(2.32)=10.67.38

or3.22to17.98

33. a.s=8.7560

b. x = 11 S( xi - x ) 2 = 180

1 ( xp - x )
2
1 (8 - 11) 2
s yp = s = 8.7560 = 4.3780
n S( xi - x ) 2
5 180

y = 68 - 3 x = 68 - 3(8) = 44

y p t / 2 s y p

443.182(4.3780)=4413.93

or30.07to57.93

1 ( xp - x )
2
1 (8 - 11)2
c. sind = s 1 = 8.7560 1 = 9.7895
n S( xi - x ) 2
5 180

1425
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

d. y p t / 2 sind

443.182(9.7895)=4431.15

or12.85to75.15

34. s=6.5141

x = 10 S( xi - x )2 = 190

1 ( xp - x )
2
1 (12 - 10) 2
s yp = s = 6.5141 = 3.0627
n S( xi - x ) 2
5 190

y = 7.6 .9 x = 7.6 .9(12) = 18.40

y p t / 2 s y p

18.403.182(3.0627)=18.409.75

or8.65to28.15

1 ( xp - x )
2
1 (12 - 10) 2
sind = s 1 = 6.5141 1 = 7.1982
n S( xi - x ) 2 5 190

y p t / 2 sind

18.403.182(7.1982)=18.4022.90

or4.50to41.30

Thetwointervalsaredifferentbecausethereismorevariabilityassociatedwithpredictingan
individualvaluethanthereisameanvalue.

35. a. Note:someofthevaluesshownwerecomputedinexercises18and26.

s=145.89

x = 3.2 S( xi - x ) 2 = 0.74

1 ( xp - x )
2
1 (3 - 3.2) 2
s yp = s = 145.89 = 68.54
n S( xi - x ) 2 6 0.74

y= 1790.5 581.1x = 1790.5 581.1(3) = 3533.8

y p t / 2 s y p

1426
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

3533.82.776(68.54)=3533.8190.27

or$3343.53to$3724.07

1 ( xp - x )
2
1 (3 - 3.2) 2
b. sind = s 1 = 145.89 1 = 161.19
n S( xi - x ) 2 6 0.74

y p t / 2 sind
3533.82.776(161.19)=3533.8447.46

or$3086.34to$3981.26

36. a. y = 359.2668 - 5.2772 x =359.2668 5.2772(30) = 200.95 or approximately $201.

b. s = 37.9372

x = 19 S( xi - x ) 2 = 1912

1 ( xp - x )
2
1 (30 - 19) 2
s yp = s = 37.9372 = 14.90
n S( xi - x ) 2
11 1912

y p t / 2 s y p

200.95 2.262 (14.90) = 200.95 33.70

or 167.25 to 234.65

1 ( xp - x )
2
1 (30 - 19) 2
c. sind = s 1 = 37.9372 1 = 40.76
n S( xi - x ) 2 11 1912

y p t / 2 sind

200.95 2.262 (40.76) = 200.95 92.20

or 108.75 to 293.15

d. As expected, the prediction interval is much wider than the confidence interval. This is due to the
fact that it is more difficult to predict the results for one new model with a temperature rating of 30
Fo than it is to estimate the mean for all models with a temperature rating of 30 Fo .

37. a. x = 57 S( xi - x ) 2 = 7648

s2=1.88s=1.37

1 ( xp - x )
2
1 (52.5 - 57) 2
s yp = s = 1.37 = 0.52
n S( xi - x ) 2 7 7648

y p t / 2 s y p

1427
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

y =4.68+0.16x=4.68+0.16(52.5)=13.08

13.082.571(.52)=13.081.34

or11.74to14.42or$11,740to$14,420

b. sind=1.47

13.082.571(1.47)=13.083.78

or9.30to16.86or$9,300to$16,860

c. Yes,$20,400ismuchlargerthananticipated.

d. Anydeductionsexceedingthe$16,860upperlimitcouldsuggestanaudit.

38. a. y = 1246.67 7.6(500) = $5046.67

b. x = 575 S( xi - x ) 2 = 93, 750

s2=MSE=58,333.33s=241.52

1 ( xp - x )
2
1 (500 - 575) 2
sind = s 1 = 241.52 1 = 267.50
n S( xi - x ) 2 6 93, 750

y p t / 2 sind

5046.674.604(267.50)=5046.671231.57

or$3815.10to$6278.24

c. Basedononemonth,$6000isnotoutoflinesince$3815.10to$6278.24isthepredictioninterval.
However,asequenceoffivetosevenmonthswithconsistentlyhighcostsshouldcauseconcern.

39. a. Letx=milesoftrackandy=weekdayridershipinthousands.

Sxi 203 Syi 309


x= = = 29 y= = = 44.1429
n 7 n 7

S( xi - x )( yi - y ) = 1471 S( xi - x ) 2 = 838

S( xi - x )( yi - y ) 1471
b1 = = = 1.7554
S( xi - x ) 2 838

b0 = y - b1 x = 44.1429 - (1.7554)(29) = -6.76

1428
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

y= -6.76 1.755 x

b. SST=3620.9SSE=1038.7SSR=2582.1

r2=SSR/SST=2582.1/3620.9=.713

Theestimatedregressionequationexplained71.3%ofthevariabilityiny;agoodfit.

c. s2=MSE=1038.7/5=207.7

s = 207.7 = 14.41

1 ( xp - x )
2
1 (30 - 29)2
s yp = s = 14.41 = 5.47
n S( xi - x ) 2
7 838

y= -6.76 1.755 x = -6.76 1.755(30) = 45.9

y p t / 2 s y p

45.92.571(5.47)=45.914.1

or31.8to60

1 ( xp - x )
2
1 (30 - 29) 2
d. sind = s 1 = 14.41 1 = 15.41
n S( xi - x ) 2
7 838

y p t / 2 sind

45.92.571(15.41)=45.939.6

or6.3to85.5

Thepredictionintervalissowidethatitwouldnotbeofmuchvalueintheplanningprocess.A
largerdatasetwouldbebeneficial.

40. a. 9

b. y=20.0+7.21x

c. 1.3626

d. SSE=SSTSSR=51,984.141,587.3=10,396.8

MSE=10,396.8/7=1,485.3

F=MSR/MSE=41,587.3/1,485.3=28.00

1429
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

UsingFtable(1degreeoffreedomnumeratorand7denominator),pvalueislessthan.01

Using Excel or Minitab, the p-value corresponding to F = 28.00 is .0011.

Becausepvalue =.05,werejectH0:B1=0.

Sellingpriceisrelatedtoannualgrossrents.

e. y=20.0+7.21(50)=380.5or$380,500

41. a. y=6.1092+.8951x

b1 - B1 .8951 - 0
b. t= = = 6.01
sb1 .149

Usingthettable(8degreesoffreedom),areaintailislessthan.005
pvalueislessthan.01

Using Excel or Minitab, the p-value corresponding to t = 6.01 is .0003.

Becausepvalue =.05,werejectH0:B1=0

Maintenanceexpenseisrelatedtousage.

c. y=6.1092+.8951(25)=28.49or$28.49permonth

42 a. y=80.0+50.0x

b. 30

c. F=MSR/MSE=6828.6/82.1=83.17

UsingFtable(1degreeoffreedomnumeratorand28denominator),pvalueislessthan.01

Using Excel or Minitab, the p-value corresponding to F = 83.17 is .000.

Becausepvalue< =.05,werejectH0:B1=0.

Annualsalesisrelatedtothenumberofsalespersons.

d. y=80+50(12)=680or$680,000

43. a. TheMinitaboutputisshownbelow:

The regression equation is


Price = 4.98 + 2.94 Weight

Predictor Coef SE Coef T P


Constant 4.979 3.380 1.47 0.154

1430
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

Weight 2.9370 0.2934 10.01 0.000

S = 8.457 R-Sq = 80.7% R-Sq(adj) = 79.9%

Analysis of Variance

Source DF SS MS F P
Regression 1 7167.9 7167.9 100.22 0.000
Residual Error 24 1716.6 71.5
Total 25 8884.5

Predicted Values for New Observations

New Obs Fit SE Fit 95.0% CI 95.0% PI


1 34.35 1.66 ( 30.93, 37.77) ( 16.56,
52.14)

b. Thepvalue=.000< =.05(torF);significantrelationship
c. r2=.807.Theleastsquareslineprovidedaverygoodfit.

d. The95%confidenceintervalis30.93to37.77.

e. The95%predictionintervalis16.56to52.14.

44. a. Scatter diagram:

b. There appears to be a negative linear relationship between the two variables. The heavier helmets
tend to be less expensive.

c. The Minitab output is shown below:

The regression equation is


Price = 2044 - 28.3 Weight

Predictor Coef SE Coef T P


Constant 2044.4 226.4 9.03 0.000
Weight -28.350 3.826 -7.41 0.000

1431
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

S = 91.8098 R-Sq = 77.4% R-Sq(adj) = 76.0%

Analysis of Variance

Source DF SS MS F P
Regression 1 462761 462761 54.90 0.000
Residual Error 16 134865 8429
Total 17 597626

y= 2044.4 28.35 Weight

d. Significant relationship: p-value = .000 < = .05

e. r2 = 0.774; A good fit

Sxi 70 Syi 76
45. a. x= = = 14 y= = = 15.2
n 5 n 5

S( xi - x )( yi - y ) = 200 S( xi - x ) 2 = 126

S( xi - x )( yi - y ) 200
b1 = = = 1.5873
S( xi - x ) 2 126

b0 = y - b1 x = 15.2 - (1.5873)(14) = -7.0222

y = -7.02 1.59 x

b. Theresidualsare3.48,2.47,4.83,1.6,and5.22

c.

2
Residuals

-2

-4

-6
0 5 10 15 20 25
x

With only 5 observations it is difficult to determine if the assumptions are satisfied.


However, the plot does suggest curvature in the residuals that would indicate that the error
term assumptions are not satisfied. The scatter diagram for these data also indicates that
the underlying relationship between x and y may be curvilinear.

d. s 2 = 23.78

1432
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

1 ( xi - x ) 2 1 ( xi - 14) 2
hi = =
n S( xi - x ) 2 5 126

The standardized residuals are 1.32, -.59, -1.11, -.40, 1.49.

e. The standardized residual plot has the same shape as the original residual plot. The
curvature observed indicates that the assumptions regarding the error term may not be
satisfied.

46. a. y = 2.32 .64 x

b.

4
3
2
Residuals

1
0
-1
-2
-3
-4
0 2 4 6 8 10
x

The assumption that the variance is the same for all values of x is questionable. The variance
appears to increase for larger values of x.

47. a. Let x = advertising expenditures and y = revenue

y = 29.4 1.55 x

b. SST = 1002 SSE = 310.28 SSR = 691.72

MSR = SSR / 1 = 691.72

MSE = SSE / (n - 2) = 310.28/ 5 = 62.0554

F = MSR / MSE = 691.72/ 62.0554= 11.15

UsingFtable(1degreeoffreedomnumeratorand5denominator),pvalueisbetween.01and.025

Using Excel or Minitab, the p-value corresponding to F = 11.15 is .0206.

Becausepvalue =.05,we conclude that the two variables are related.

1433
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

c.

10

5
Residuals

-5

-10

-15
25 30 35 40 45 50 55 60 65
Predicted Values

d. The residual plot leads us to question the assumption of a linear relationship between x and y. Even
though the relationship is significant at the .05 level of significance, it would be extremely
dangerous to extrapolate beyond the range of the data.

48. a. y = 80 4 x

1434
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

2
Residuals
0

-2

-4

-6

-8
0 2 4 6 8 10 12 14
x

b. The assumptions concerning the error term appear reasonable.

49. a. The Minitab output follows:

The regression equation is


Price ($) = 22636 + 59.0 Square Footage

Predictor Coef SE Coef T P


Constant 22636 20460 1.11 0.283
Square Footage 58.96 12.08 4.88 0.000

S = 19166.0 R-Sq = 57.0% R-Sq(adj) = 54.6%

Analysis of Variance

Source DF SS MS F P
Regression 1 8748562231 8748562231 23.82 0.000
Residual Error 18 6612039769 367335543
Total 19 15360602000

b.

1435
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

c. The residual plot leads us to question the assumption of a linear relationship between square
footage and price. Therefore, even though the relationship is very significant (p-value = .000),
using the estimated regression equation make predictions of the price for a house with square
footage beyond the range of the data is not recommended.

50. a. The Minitab output follows:

Theregressionequationis
Y=66.1+0.402X

PredictorCoefSECoefTp
Constant66.1032.062.060.094
X0.40230.22761.770.137

S=12.62Rsq=38.5%Rsq(adj)=26.1%

AnalysisofVariance

SOURCEDFSSMSFp
Regression1497.2497.23.120.137
ResidualError5795.7159.1
Total61292.9

UnusualObservations
Obs.XYFitSEFitResidualSt.Resid
1135145.00120.424.8724.582.11R

Rdenotesanobservationwithalargestandardizedresidual.

The standardized residuals are: 2.11, -1.08, .14, -.38, -.78, -.04, -.41

The first observation appears to be an outlier since it has a large standardized residual.

b.

1436
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

2.5

2.0

1.5

Standardized Residual 1.0

0.5

0.0

-0.5

-1.0

110 115 120 125 130 135 140


Fitted Value

Thestandardizedresidualplotindicatesthattheobservationx=135,y=145maybeanoutlier;
notethatthisobservationhasastandardizedresidualof2.11.

c. Thescatterdiagramisshownbelow

1437
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

Thescatterdiagramalsoindicatesthattheobservationx=135,y=145maybeanoutlier;the
implicationisthatforsimplelinearregressionanoutliercanbeidentifiedbylookingatthescatter
diagram.

51. a. TheMinitaboutputisshownbelow:

Theregressionequationis
Y=13.0+0.425X

PredictorCoefSECoefTp
Constant13.0022.3965.430.002
X0.42480.21162.010.091

S=3.181Rsq=40.2%Rsq(adj)=30.2%

AnalysisofVariance

SOURCEDFSSMSFp
Regression140.7840.784.030.091
ResidualError660.7210.12
Total7101.50

UnusualObservations
Obs.XYFitStdev.FitResidualSt.Resid
712.024.0018.101.205.902.00R
822.019.0022.352.783.352.16RX

Rdenotesanobservationwithalargestandardizedresidual.
XdenotesanobservationwhoseXvaluegivesitlargeinfluence.

Thestandardizedresidualsare:1.00,.41,.01,.48,.25,.65,2.00,2.16

Thelasttwoobservationsinthedatasetappeartobeoutlierssincethestandardizedresidualsfor
theseobservationsare2.00and2.16,respectively.

1438
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

b. UsingMinitab,weobtainedthefollowingleveragevalues:

.28,.24,.16,.14,.13,.14,.14,.76

MINITABidentifiesanobservationashavinghighleverageifhi>6/n;forthesedata,6/n=
6/8=.75.Sincetheleveragefortheobservationx=22,y=19is.76,Minitabwouldidentify
observation8asahighleveragepoint.Thus,weconcludethatobservation8isaninfluential
observation.

c.

Thescatterdiagramindicatesthattheobservationx=22,y=19isaninfluentialobservation.

52. a. TheMinitaboutputisshownbelow:

The regression equation is


Shipment = 4.09 + 0.196 Media$

Predictor Coef SE Coef T P


Constant 4.089 2.168 1.89 0.096
MediaExp 0.19552 0.03635 5.38 0.001

S = 5.044 R-Sq = 78.3% R-Sq(adj) = 75.6%

Analysis of Variance

Source DF SS MS F P
Regression 1 735.84 735.84 28.93 0.001
Residual Error 8 203.51 25.44
Total 9 939.35

Unusual Observations
Obs Media$ Shipment Fit SE Fit Residual St Resid
1 120 36.30 27.55 3.30 8.75 2.30R

1439
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

R denotes an observation with a large standardized residual

b. Minitabidentifiesobservation1ashavingalargestandardizedresidual;thus,wewouldconsider
observation1tobeanoutlier.

53. a. TheMinitaboutputisshownbelow:

The regression equation is


Price = 28.0 + 0.173 Volume

Predictor Coef SE Coef T P


Constant 27.958 4.521 6.18 0.000
Volume 0.17289 0.07804 2.22 0.036

S = 17.53 R-Sq = 17.0% R-Sq(adj) = 13.5%

Analysis of Variance

Source DF SS MS F P
Regression 1 1508.4 1508.4 4.91 0.036
Residual Error 24 7376.1 307.3
Total 25 8884.5

Unusual Observations
Obs Volume Price Fit SE Fit Residual St Resid
22 230 40.00 67.72 15.40 -27.72
-3.31RX

R denotes an observation with a large standardized residual


X denotes an observation whose X value gives it large influence.

b. The Minitab output identifies observation 22 as having a large standardized residual and is an
observation whose x value gives it large influence. The following residual plot verifies these
observations.

1
Standardized Residual

-1

-2

-3

-4
30 40 50 60 70
Fitted Value

1440
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

54. a.

The scatter diagram does indicate potential outliers and/or influential observations. For example,
the data for the Washington Redskins, New England Patriots, and the Dallas Cowboys not only
have the three highest revenues, they also have the highest team values.

b. A portion of the Minitab output follows:

The regression equation is


Value = - 252 + 5.83 Revenue

Predictor Coef SE Coef T P


Constant -252.1 130.8 -1.93 0.064
Revenue 5.8317 0.5863 9.95 0.000

S = 87.2441 R-Sq = 76.7% R-Sq(adj) = 76.0%

Analysis of Variance

Source DF SS MS F P
Regression 1 753008 753008 98.93 0.000
Residual Error 30 228346 7612
Total 31 981354

Unusual Observations

Obs Revenue Value Fit SE Fit Residual St Resid


9 269 1612.0 1316.6 31.8 295.4 3.64R
19 282 1324.0 1392.5 38.6 -68.5 -0.88 X
21 214 1178.0 995.9 16.0 182.1 2.12R
22 213 1170.0 990.1 16.2 179.9 2.10R
32 327 1538.0 1654.9 63.7 -116.9 -1.96 X

R denotes an observation with a large standardized residual.


X denotes an observation whose X value gives it large leverage.

1441
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

c. The Minitab output indicates that there are five unusual observations:

Observation 9 (Dallas Cowboys) is an outlier because it has a large standardized residual.

Observation 19 (New England Patriots) is an influential observation becasuse has high leverage.

Observation 21 (New York Giants) is an outlier because it has a large standardized residual.

Observation 22 (New York Jets) is an outlier because it has a large standardized residual.

Observation 32 (Washington Redskins) is an influential observation becasuse has high leverage.

55. No.Regressionorcorrelationanalysiscanneverprovethattwovariablesarecausallyrelated.

56. Theestimateofameanvalueisanestimateoftheaverageofallyvaluesassociatedwiththesame
x.Theestimateofanindividualyvalueisanestimateofonlyoneoftheyvaluesassociatedwitha
particularx.

57. Thepurposeoftestingwhether 1 = 0 istodeterminewhetherornotthereisasignificant


relationshipbetweenxandy.However,rejecting 1 = 0 doesnotnecessarilyimplyagoodfit.For
example,if 1 = 0 isrejectedandr2islow,thereisastatisticallysignificantrelationshipbetweenx
andybutthefitisnotverygood.

58. a. The Minitab output is shown below:

The regression equation is


Price = 9.26 + 0.711 Shares

Predictor Coef SE Coef T P


Constant 9.265 1.099 8.43 0.000
Shares 0.7105 0.1474 4.82 0.001

S = 1.419 R-Sq = 74.4% R-Sq(adj) = 71.2%

Analysis of Variance

Source DF SS MS F P
Regression 1 46.784 46.784 23.22 0.001
Residual Error 8 16.116 2.015
Total 9 62.900

b. Since the p-value corresponding to F = 23.22 = .001 < = .05, the relationship is significant.

c. r 2 = .744; a good fit. The least squares line explained 74.4% of the variability in Price.

d. y = 9.26 .711(6) = 13.53

59. a. The Minitab output is shown below:

The regression equation is


Share Price ($) = - 2.99 + 0.911 Fair Value ($)

Predictor Coef SE Coef T P


Constant -2.987 5.791 -0.52 0.610
Fair Value ($) 0.91128 0.09783 9.31 0.000

1442
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

S = 12.0064 R-Sq = 76.9% R-Sq(adj) = 76.1%

Analysis of Variance

Source DF SS MS F P
Regression 1 12507 12507 86.76 0.000
Residual Error 26 3748 144
Total 27 16255

y= -2.987 + .91128 Fair Value ($)

b. Significant relationship: p-value = .000 < = .05

c. y= -2.987 + .91128 Fair Value ($) = -2.987 + .91128(50) = 42.577 or approximately $42.58

d. The estimated regression equation should provide a good estimate because r2 = 0.769

60. a.

The scatter diagram indicates a positive linear relationship between the two variables. Online
universities with higher retention rates tend to have higher graduation rates.

b. The Minitab output follows:

The regression equation is


GR(%) = 25.4 + 0.285 RR(%)

Predictor Coef SE Coef T P


Constant 25.423 3.746 6.79 0.000
RR(%) 0.28453 0.06063 4.69 0.000

S = 7.45610 R-Sq = 44.9% R-Sq(adj) = 42.9%

Analysis of Variance

Source DF SS MS F P

1443
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

Regression 1 1224.3 1224.3 22.02 0.000


Residual Error 27 1501.0 55.6
Total 28 2725.3

Unusual Observations

Obs RR(%) GR(%) Fit SE Fit Residual St Resid


2 51 25.00 39.93 1.44 -14.93 -2.04R
3 4 28.00 26.56 3.52 1.44 0.22 X

R denotes an observation with a large standardized residual.


X denotes an observation whose X value gives it large leverage.

c. Because the p-value = .000 < =.05, the relationship is significant.

d. The estimated regression equation is able to explain 44.9% of the variability in the graduation rate
based upon the linear relationship with the retention rate. It is not a great fit, but given the type of
data, the fit is reasonably good.

e. In the Minitab output in part (b), South University is identified as an observation with a large
standardized residual. With a retention rate of 51% it does appear that the graduation rate of 25% is
low as compared to the results for other online universities. The president of South University
should be concerned after looking at the data. Using the estimated regression equation, we estimate
that the gradation rate at South University should be 25.4 + .285(51) = 40%.

f. In the Minitab output in part (b), the University of Phoenix is identified as an observation whose x
value gives it large influence. With a retention rate of only 4%, the president of the University of
Phoenix should be concerned after looking at the data.

61. TheMinitaboutputisshownbelow:

Theregressionequationis
Expense=10.5+0.953Usage

PredictorCoefSECoefTp
Constant10.5283.7452.810.023
X0.95340.13826.900.000

S=4.250Rsq=85.6%Rsq(adj)=83.8%

AnalysisofVariance

SOURCEDFSSMSFp
Regression1860.05860.0547.620.000
ResidualError8144.4718.06
Total91004.53

FitStdev.Fit95%C.I.95%P.I.
39.131.49(35.69,42.57)(28.74,49.52)

a. y=10.528+.9534Usage

b. SincethepvaluecorrespondingtoF=47.62=.000<=.05,werejectH0: 1=0.

c. The95%predictionintervalis28.74to49.52or$2874to$4952

1444
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

d. Yes,sincetheexpectedexpenseis y=10.528+.9534(30)=39.13or$3913.

62. a. The Minitab output is shown below:

The regression equation is


Defects = 22.2 - 0.148 Speed

Predictor Coef SE Coef T P


Constant 22.174 1.653 13.42 0.000
Speed -0.14783 0.04391 -3.37 0.028

S = 1.489 R-Sq = 73.9% R-Sq(adj) = 67.4%

Analysis of Variance

Source DF SS MS F P
Regression 1 25.130 25.130 11.33 0.028
Residual Error 4 8.870 2.217
Total 5 34.000

Predicted Values for New Observations

New Obs Fit SE Fit 95.0% CI 95.0% PI


1 14.783 0.896 ( 12.294, 17.271) ( 9.957, 19.608)

b. Since the p-value corresponding to F = 11.33 = .028 < = .05, the relationship is significant.

c. r 2 = .739; a good fit. The least squares line explained 73.9% of the variability in the number of
defects.

d. Using the Minitab output in part (a), the 95% confidence interval is 12.294 to 17.271.

63. a.

1445
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

Thereappearstobeanegativelinearrelationshipbetweendistancetoworkandnumberofdays
absent.

b. TheMinitaboutputisshownbelow:

Theregressionequationis
Days=8.100.344Distance

PredictorCoefSECoefTp
Constant8.09780.808810.010.000
X0.344200.077614.430.002

S=1.289Rsq=71.1%Rsq(adj)=67.5%

AnalysisofVariance

SOURCEDFSSMSFp
Regression132.69932.69919.670.002
ResidualError813.3011.663
Total946.000

FitStdev.Fit95%C.I.95%P.I.
6.3770.512(5.195,7.559)(3.176,9.577)

c. SincethepvaluecorrespondingtoF=419.67is.002<=.05.WerejectH0: 1=0.

Thereisasignificantrelationshipbetweenthenumberofdaysabsentandthedistancetowork.

d. r2=.711.Theestimatedregressionequationexplained71.1%ofthevariabilityiny;thisisa
reasonablygoodfit.

e. The95%confidenceintervalis5.195to7.559orapproximately5.2to7.6days.

1446
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

64. a. TheMinitaboutputisshownbelow:

Theregressionequationis
Cost=220+132Age

PredictorCoefSECoefTp
Constant220.0058.483.760.006
X131.6717.807.400.000

S=75.50Rsq=87.3%Rsq(adj)=85.7%

AnalysisofVariance

SOURCEDFSSMSFp
Regression131205031205054.750.000
ResidualError8456005700
Total9357650

FitStdev.Fit95%C.I.95%P.I.
746.729.8(678.0,815.4)(559.5,933.9)

b. SincethepvaluecorrespondingtoF=54.75is.000<=.05,werejectH0:1=0.

Maintenancecostandageofbusarerelated.

c. r2=.873.Theleastsquareslineprovidedaverygoodfit.

d. The95%predictionintervalis559.5to933.9or$559.50to$933.90

65. a. TheMinitaboutputisshownbelow:

Theregressionequationis
Points=5.85+0.830Hours

PredictorCoefSECoefTp
Constant5.8477.9720.730.484
X0.82950.10957.580.000

S=7.523Rsq=87.8%Rsq(adj)=86.2%

AnalysisofVariance

SOURCEDFSSMSFp
Regression13249.73249.757.420.000
ResidualError8452.856.6
Total93702.5

FitStdev.Fit95%C.I.95%P.I.
84.653.67(76.19,93.11)(65.35,103.96)

b. SincethepvaluecorrespondingtoF=57.42is.000<=.05,werejectH0:1=0.

Totalpointsearnedisrelatedtothehoursspentstudying.

c. 84.65points

d. The95%predictionintervalis65.35to103.96

1447
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

66. a. TheMinitaboutputisshownbelow:

Theregressionequationis
Horizon=0.275+0.950S&P500

PredictorCoefSECoefTP
Constant0.27470.90040.310.768
S&P5000.94980.35692.660.029

S=2.664RSq=47.0%RSq(adj)=40.3%

AnalysisofVariance

SourceDFSSMSFP
Regression150.25550.2557.080.029
ResidualError856.7817.098
Total9107.036

ThemarketbetaforHorizonisb1=.95

b. Sincethepvalue=0.029islessthan=.05,therelationshipissignificant.

c. r2=.470.Theleastsquareslinedoesnotprovideaverygoodfit.

d. Xeroxhashigherriskwithamarketbetaof1.22.

67. a. TheMinitaboutputisshownbelow:

The regression equation is


Audit% = - 0.471 +0.000039 Income

Predictor Coef SE Coef T P


Constant -0.4710 0.5842 -0.81 0.431
Income 0.00003868 0.00001731 2.23 0.038

S = 0.2088 R-Sq = 21.7% R-Sq(adj) = 17.4%

Analysis of Variance

Source DF SS MS F P
Regression 1 0.21749 0.21749 4.99 0.038
Residual Error 18 0.78451 0.04358
Total 19 1.00200

Predicted Values for New Observations

New Obs Fit SE Fit 95.0% CI 95.0% PI


1 0.8828 0.0523 ( 0.7729, 0.9927) ( 0.4306, 1.3349)

b. Sincethepvalue=0.038islessthan=.05,therelationshipissignificant.

c. r2=.217.Theleastsquareslinedoesnotprovideaverygoodfit.

1448
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
SimpleLinearRegression

d. The95%confidenceintervalis.7729to.9927.

68. a.

b. There appears to be a positive linear relationship between the two variables.

c. A portion of the Minitab output for this problem follows.

The regression equation is


Rating (%) = 9.4 + 1.29 Top Five (%)

Predictor Coef SE Coef T P


Constant 9.37 10.24 0.92 0.380
Top Five (%) 1.2875 0.2293 5.61 0.000

S = 6.62647 R-Sq = 74.1% R-Sq(adj) = 71.8%

Analysis of Variance

Source DF SS MS F P
Regression 1 1383.9 1383.9 31.52 0.000
Residual Error 11 483.0 43.9
Total 12 1866.9

y = 9.37 + 1.2875 Top Five (%)

d. Since the p-value = .000 < = .05, the relationship is significant.

1449
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Chapter14

e. r 2 = .741; a good fit. The least squares line explained 74.1% of the variability in the satisfaction
rating.

f. rxy = r 2 = .741 = .86

1450
2010 Cengage Learning. All Rights Reserved.
May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.