SBE_SM14.doc
-
Upload
wilson-quispe-condori -
Category
Documents
-
view
220 -
download
0
Transcript of SBE_SM14.doc
-
8/13/2019 SBE_SM14.doc
1/44
Chapter 14
Simple Linear Regression
Learning Objectives
1. Understand how regression analysis can be used to develop an equation that estimatesmathematically how two variables are related.
2. Understand the differences between the regression model, the regression equation, and theestimated regression equation.
3. Know how to fit an estimated regression equation to a set of sample data based upon the least-squares method.
4. Be able to determine how good a fit is provided by the estimated regression equation and computethe sample correlation coefficient from the regression analysis output.
5. Understand the assumptions necessary for statistical inference and be able to test for a significantrelationship.
6. Know how to develop confidence interval estimates ofygiven a specific value ofxin both the caseof a mean value ofyand an individual value ofy.
7. Learn how to use a residual plot to make a judgement as to the validity of the regressionassumptions, recognize outliers, and identify influential observations.
8. Know the definition of the following terms:
independent and dependent variablesimple linear regressionregression modelregression equation and estimated regression equationscatter diagramcoefficient of determinationstandard error of the estimateconfidence intervalprediction intervalresidual plotstandardized residual plotoutlierinfluential observation
leverage
14 - 1
-
8/13/2019 SBE_SM14.doc
2/44
Chapter 14
Solutions:
1 a.
b. There appears to be a linear relationship betweenx andy.
c. Many different straight lines can be drawn to provide a linear approximation of therelationship betweenx and y; in part d we will determine the equation of a straight linethat best represents the relationship according to the least squares criterion.
d. Summations needed to compute the slope and y-intercept are:
215 40 ( )( ) 26 ( ) 10i i i i ix y x x y y x x = = = =
1 2
( )( ) 262.6
10( )
i i
i
x x y yb
x x
= = =
b y b x0 1 8 2 6 3 0 2= = =( . )( ) .
0.2 2.6y x= +
e. 0.2 2.6(4) 10.6y= + =
14 - 2
0
2
4
6
8
10
1214
16
0 1 2 3 4 5 6
x
y
-
8/13/2019 SBE_SM14.doc
3/44
Simple Linear Regression
2. a.
b. There appears to be a linear relationship betweenx andy.
c. Many different straight lines can be drawn to provide a linear approximation of therelationship betweenx and y; in part d we will determine the equation of a straight line thatbest represents the relationship according to the least squares criterion.
d. Summations needed to compute the slope andy-intercept are:
219 116 ( )( ) 57.8 ( ) 30.8i i i i ix y x x y y x x = = = =
1 2
( )( ) 57.81.8766
30.8( )
i i
i
x x y yb
x x
= = =
b y b x0 1 232 18766 38 30 3311= = =. ( . )( . ) .
. .y x= 30 33 188
e. . . ( ) .y= =30 33 188 6 19 05
14 - 3
0
5
10
15
20
25
30
35
0 1 2 3 4 5 6 7 8 9
x
y
-
8/13/2019 SBE_SM14.doc
4/44
Chapter 14
3. a.
b. Summations needed to compute the slope andy-intercept are:
226 17 ( )( ) 11.6 ( ) 22.8i i i i ix y x x y y x x = = = =
1 2
( )( ) 11.60.5088
22.8( )
i i
i
x x y yb
x x
= = =
b y b x0 1 34 05088 2 0 7542= = =. ( . )(5. ) .
. .y x= +0 75 0 51
c. . . ( ) .y= + =0 75 0 51 4 2 79
14 - 4
0
1
2
3
45
6
7
0 1 2 3 4 5 6 7 8 9
x
y
-
8/13/2019 SBE_SM14.doc
5/44
100
105
110
115
120
125
130
135
61 62 63 64 65 66 67 68 69
x
y
Simple Linear Regression
4. a.
b.There appears to be a linear relationship betweenx andy.
c. Many different straight lines can be drawn to provide a linear approximation of therelationship betweenx and y; in part d we will determine the equation of a straight line thatbest represents the relationship according to the least squares criterion.
d. Summations needed to compute the slope andy-intercept are:
2325 585 ( )( ) 110 ( ) 20i i i i ix y x x y y x x = = = =
1 2
( )( ) 1105.5
20( )
i i
i
x x y yb
x x
= = =
b y b x0 1 117 5 65 2405= = = (5. )( ) .
. .y x= +2405 55
e. . . . . ( )y x= + = + =2405 55 2405 55 63 106 pounds
14 - 5
-
8/13/2019 SBE_SM14.doc
6/44
Chapter 14
5. a.
0
500
1000
1500
2000
2500
1 1.5 2 2.5 3 3.5 4 4.5
Baggage Capacity
Price($)
b. Letx= baggage capacity andy= price ($).
There appears to be a linear relationship betweenx andy.
c. Many different straight lines can be drawn to provide a linear approximation of therelationship betweenx andy; in part (d) we will determine the equation of a straight linethat best represents the relationship according to the least squares criterion.
d. Summations needed to compute the slope andy-intercept are:
229.5 11,110 ( )( ) 2909.8889 ( ) 4.5556i i i i ix y x x y y x x = = = =
1 2
( )( ) 2909.9889638.7561
4.5556( )
i i
i
x x y yb
x x
= = =
0 1 1234.4444 (638.7561)(3.2778) 859.26b y b x= = =
859.26 638.76y x= +
e. A one point increase in the baggage capacity rating will increase the price by approximately $639.
f. 859.26 638.76 859.26 638.76(3) $1057y x= + = + =
14 - 6
-
8/13/2019 SBE_SM14.doc
7/44
Simple Linear Regression
6. a.
0
5
10
15
20
25
30
50.00 70.00 90.00 110.00 130.00 150.00 170.00 190.00
Salary
Bonus
b. Letx= salary andy= bonus.
There appears to be a linear relationship betweenx andy.
c. Summations needed to compute the slope andy-intercept are:
21420 160 ( )( ) 1114 ( ) 6046i i i i ix y x x y y x x = = = =
1 2
( )( ) 1114.18425
6046( )
i i
i
x x y yb
x x
= = =
0 1 16 (.18425)(142) 10.16b y b x= = =
10.16 .18y x
= +
d. $1000 increase in salary will increase the bonus by $180.
e 10.16 .18 10.16 .18(120) 11.95y x= + = + = or approximately $12,000
14 - 7
-
8/13/2019 SBE_SM14.doc
8/44
Chapter 14
7. a.
b. Letx= DJIA andy= S&. S!""at#n% needed t "'!te te %'e and y*#nte+e't a+e
2104,850 14,233 ( )( ) 268,921 ( ) 1,806,384i i i i ix y x x y y x x = = = =
1 2
( )( ) 268,9210.14887
1,806,384( )
i i
i
x x y yb
x x
= = =
0 1 1423.3 (.14887)(10,485) 137.629b y b x= = =
137.63 0.1489y x= +
. 137.63 0.1489(11,000) 1500.27y= + = + a''+-#"ate 1500
8. a.
14 - 8
1300
1350
1400
1450
1500
1550
9600 9800 10000 10200 10400 10600 10800 11000 11200
DJIA
S&
-
8/13/2019 SBE_SM14.doc
9/44
Simple Linear Regression
0
200
400
600
800
1000
1200
1400
1600
1800
0 1 2 3 4 5 6
Sidetrack Capability
Price
b. /e+e a''ea+% t be a #nea+ +eat#n%#' beteenx = %#det+a a'ab##t andy = '+#e, #t #e+'+#ed "de% a#n a #e+ ee and#n.
. S!""at#n% needed t "'!te te %'e and *#nte+e't a+e
228 10,621 ( )( ) 4003.2 ( ) 19.6i i i i ix y x x y y x x = = = =
1 2
( )( ) 4003.2204.2449
19.6( )
i i
i
x x y yb
x x
= = =
0 1 1062.1 (204.2449)(2.8) 490.21b y b x= = =
490.21 204.24y x= +
d. 490.21 204.24 490.21 204.24(4) 1307y x= + = + =
9. a.
14 - 9
-
8/13/2019 SBE_SM14.doc
10/44
Chapter 14
b. Summations needed to compute the slope andy-intercept are:
270 1080 ( )( ) 568 ( ) 142i i i i ix y x x y y x x = = = =
1 2
( )( ) 5684
142( )
i i
i
x x y yb
x x
= = =
b y b x0 1 108 4 7 80= = =( )( )
y x= +80 4
c. ( )y x= + = + =80 4 80 4 9 116
14 - 10
50
60
70
80
90
100
110
120
130
140
150
0 2 4 6 8 10 12 14x
y
-
8/13/2019 SBE_SM14.doc
11/44
Simple Linear Regression
10. a.
b. Letx= 'e++"ane %+e andy= e+a +at#n. S!""at#n% needed t "'!te te %'e and y*#nte+e't a+e
22752 1177 ( )( ) 1723.73 ( ) 11,867.73i i i i ix y x x y y x x = = = =
1 2
( )( ) 1723.730.1452
11,867.73( )
i i
i
x x y yb
x x
= = =
0 1 78.4667 (.1452)(183.4667) 51.82b y b x= = =
51.82 0.145y x= +
. 51.82 0.145(225) 84.4y= + = + a''+-#"ate 84
14 - 11
60
65
70
75
80
85
90
95
100 120 140 160 180 200 220 240 260
e+$+"an&e S&$+e
53e+a))6at#n2
-
8/13/2019 SBE_SM14.doc
12/44
Chapter 14
11. a.
0
5
10
15
20
25
30
35
10 15 20 25 30 35
Arrivals (% )
Departures(%
)
b. There appears to be a linear relationship between the variables.
c. Letx= percentage of late arrivals andy= percentage of late departures.
The summations needed to compute the slope and they-intercept are:
2273 265 ( )( ) 101 ( ) 114.4i i i i ix y x x y y x x = = = =
1 2
( )( ) 101.8829
114.4( )
i i
i
x x y yb
x x
= = =
0 1 21 (.8829)(21.4) 2.106b y b x= = =
2.106 .883y x= +
d. A one percent increase in the percentage of late arrivals will increase the percentage of late arrivalsby .88 or slightly less than one percent.
e. 2.106 .883 2.106 .883(22) 21.53y x= + = + =
14 - 12
-
8/13/2019 SBE_SM14.doc
13/44
Simple Linear Regression
12. a.
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
3,000,000
3,500,000
4,000,000
4,500,000
5,000,000
0 200 400 600 800 1,000 1,200 1,400 1,600 1,800
!"be+ $ "'$ee%
+$d!t#$n
b. /e+e a''ea+% t be a '%#t#e #nea+ +eat#n%#' beteen te n!"be+ e"'ee% and te'+d!t#n !t'!t.
. Letx= n!"be+ e"'ee% andy= '+d!t#n !t'!t. S!""at#n% needed t "'!te te %'eandy*#nte+e't a+e
258,990,16640,6 == ii yx
27272,894,161.)(499.81828,166,725,))((2== xxyyxx iii
7935.821,22727.161,894,2
8182.499,725,166,8
)(
))((
21 ==
=
xx
yyxx
bi
ii
479,384.180*(301.8182)2,821.7935*45772,284.4510 === xbyb
xy 2,821.793504*79,384.18 +=
d. y -79,834.1804 + 2,821.7935 (400) = 1,048,883
13. a.
14 - 13
-
8/13/2019 SBE_SM14.doc
14/44
Chapter 14
b. The summations needed to compute the slope and they-intercept are:
2399 97.1 ( )( ) 1233.7 ( ) 7648i i i i ix y x x y y x x = = = =
1 2
( )( ) 1233.70.16131
7648( )
i i
i
x x y yb
x x
= = =
b y b x0 1 1387143 016131 4 67675= = =. ( . )(57) .
. .y x= +4 68 016
c. . . . . (52. ) .y x= + = + =4 68 016 4 68 016 5 1308 or approximately $13,080.
The agent's request for an audit appears to be justified.
14 - 14
0.0
5.0
10.0
15.0
20.0
25.0
30.0
0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0
x
y
-
8/13/2019 SBE_SM14.doc
15/44
Simple Linear Regression
14. a.
0.0
5.0
10.0
15.0
20.0
25.0
30.0
0 100 200 300 400 500 600 700 800 900
Quantity of the house
VacancyRate(%)
b. Letx= number of houses andy= vacancy rate
The summations needed to compute the slope and they-intercept are:
62,592)(98.862,5))((312697,42==== xxyyxxyx iiiii
0099.620,592
98.862,5
)(
))((21 =
=
=
xx
yyxxb
i
ii
11.21)294.276(0099.376.1810 =+== xbyb
xy 0099.11.21 =
c. 13.19)200(0099.11.21 ==y
15. a. The estimated regression equation and the mean for the dependent variable are:
. .y x yi i= + =0 2 2 6 8
The sum of squares due to error and the total sum of squares are
SS SS/= = = =( ) . ( )y y y yi i i2 212 40 80
Thus, SSR = SST - SSE = 80 - 12.4 = 67.6
b. r2= SSR/SST = 67.6/80 = .845
The least squares line provided a very good fit; 84.5% of the variability inyhas been explained bythe least squares line.
c. r= = +. .845 9192
14 - 15
-
8/13/2019 SBE_SM14.doc
16/44
Chapter 14
16. a. The estimated regression equation and the mean for the dependent variable are:
30.33 1.88 23.2iy x y= =
The sum of squares due to error and the total sum of squares are
2 2SS ( ) 6.33 SS/ ( ) 114.80i i iy y y y= = = =
Thus, SSR = SST - SSE = 114.80 - 6.33 = 108.47
b. r2= SSR/SST = 108.47/114.80 = .945
The least squares line provided an excellent fit; 94.5% of the variability inyhas been explained bythe estimated regression equation.
c. r= = . .945 9721
Note: the sign for ris negative because the slope of the estimated regression equation is negative.
(b1= -1.88)
17. The estimated regression equation and the mean for the dependent variable are:
.75 .51 3.4iy x y= + =
The sum of squares due to error and the total sum of squares are
2 2SS ( ) 5.3 SS/ ( ) 11.2i i iy y y y= = = =
Thus, SSR = SST - SSE = 11.2 - 5.3 = 5.9
r2= SSR/SST = 5.9/11.2 = .527
We see that 52.7% of the variability inyhas been explained by the least squares line.
r= = +. .527 7259
18. a. The estimated regression equation and the mean for the dependent variable are:
1790.5 581.1 3650y x y= + =
The sum of squares due to error and the total sum of squares are
2 2SS ( ) 85,135.14 SS/ ( ) 335,000i i iy y y y= = = =
Thus, SSR = SST - SSE = 335,000 - 85,135.14 = 249,864.86
b. r2= SSR/SST = 249,864.86/335,000 = .746
We see that 74.6% of the variability inyhas been explained by the least squares line.
c. r= = +. .746 8637
19. a. The estimated regression equation and the mean for the dependent variable are:
14 - 16
-
8/13/2019 SBE_SM14.doc
17/44
Simple Linear Regression
137.63 .1489 1423.3y x y= + =
The sum of squares due to error and the total sum of squares are
2 2SS ( ) 7547.14 SS/ ( ) 47,582.10i i iy y y y= = = =
Thus, SSR = SST - SSE = 47,582.10 - 7547.14 = 40,034.96
b. r2= SSR/SST = 40,034.96/47,582.10 = .84
We see that 84% of the variability inyhas been explained by the least squares line.
c. .84 .92r= = +
20. a. Letx= #n"e andy= "e '+#e. S!""at#n% needed t "'!te te %'e andy*#nte+e't a+e
074,163)(242,41))((752,2706,142==== xxyyxxyx iiiii
253.074,163
242,41
)(
))((21 ==
=
xx
yyxxb
i
ii
6.80)225,1)(2529.0(3333.22910 === xbyb
xy 253.6.80 +=
b. The sum of squares due to error and the total sum of squares are
==== 6667.466,18)(2960.036,8)(22
yySSTyySSE iii
Thus, SSR = SST - SSE = 18,466.6667 8,036.2960 = 10,430.3707
r2= SSR/SST = 10,430.3707/18,466.6667 = .5648
We see that 56.48% of the variability inyhas been explained by the least squares line.
7515.5648. ==r
. 75.159)950(253.6.80 =+=y
21. a. The summations needed in this problem are:
23450 33,700 ( )( ) 712,500 ( ) 93,750i i i i ix y x x y y x x = = = =
1 2
( )( ) 712,5007.6
93,750( )
i i
i
x x y yb
x x
= = =
0 1 5616.67 (7.6)(575) 1246.67b y b x= = =
. .y x= +1246 67 7 6
14 - 17
-
8/13/2019 SBE_SM14.doc
18/44
Chapter 14
b. $7.60
c. The sum of squares due to error and the total sum of squares are:
2 2SS ( ) 233,333.33 SS/ ( ) 5,648,333.33i i iy y y y= = = =
Thus, SSR = SST - SSE = 5,648,333.33 - 233,333.33 = 5,415,000
r2= SSR/SST = 5,415,000/5,648,333.33 = .9587
We see that 95.87% of the variability inyhas been explained by the estimated regression equation.
d. . . . . (500) $5046.y x= + = + =1246 67 7 6 1246 67 7 6 67
22. a. Letx= speed (ppm) andy= price ($)
The summations needed in this problem are:
2
149.9 10,221 ( )( ) 34,359.81 ( ) 291.389i i i i ix y x x y y x x = = = =
1 2
( )( ) 34,359.81117.9173
291.389( )
i i
i
x x y yb
x x
= = =
0 1 1022.1 (117.9173)(14.99) 745.80b y b x= = =
745.80 117.917y x= +
b. The sum of squares due to error and the total sum of squares are:
2 2SS ( ) 1,678, 294 SS/ ( ) 5,729,911i i iy y y y= = = =
Thus, SSR = SST - SSE = 5,729,911 - 1,678,294 = 4,051,617
r2= SSR/SST = 4,051,617/5,729,911 = 0.7071
Approximately 71% of the variability in price is explained by the speed.
c. .7071 .84r= = +
It reflects a linear relationship that is between weak and strong.
23. a. s2 = MSE = SSE / (n- 2) = 12.4 / 3 = 4.133
b. s= = =:S 4133 2 033. .
c. 2( ) 10ix x =
1 2
2.0330.643
10( )b
i
ss
x x= = =
14 - 18
-
8/13/2019 SBE_SM14.doc
19/44
Simple Linear Regression
d. t b
sb= = =1
1
2 6
6434 04
.
..
Using ttable (3 degrees of freedom), area in tail is between .01 and .025
p-value is between .02 and .05
Actualp-value = .0272
Becausep-value , we rejectH0: 1= 0
e. MSR = SSR / 1 = 67.6
F = MSR / MSE = 67.6 / 4.133 = 16.36
Using Ftable (1 degree of freedom numerator and 3 denominator),p-value is between .025 and .05
Actualp-value = .0272
Becausep-value , we rejectH0: 1= 0
Source of Variation Sum of Squares Degrees of Freedom Mean Square F
Regression 67.6 1 67.6 16.36
Error 12.4 3 4.133
Total 80.0 4
24. a. s2 = MSE = SSE / (n- 2) = 6.33 / 3 = 2.11
b. s= = =:S 211 1453. .
c. 2( ) 30.8ix x =
1 2
1.4530.262
30.8( )b
i
ss
x x= = =
d. t b
sb= =
= 1
1
188
262718
.
..
Using ttable (3 degrees of freedom), area in tail is less than .005;p-value is less than .01
Actualp-value = .0056
Becausep-value
, we rejectH0: 1= 0
e. MSR = SSR / 1 = 8.47
F = MSR / MSE = 108.47 / 2.11 = 51.41
Using Ftable (1 degree of freedom numerator and 3 denominator),p-value is less than .01
Actualp-value = .0056
14 - 19
-
8/13/2019 SBE_SM14.doc
20/44
Chapter 14
Becausep-value , we rejectH0: 1= 0
Source of Variation Sum of Squares Degrees of Freedom Mean Square F
Regression 108.47 1 108.47 51.41
Error 6.33 3 2.11
Total 114.80 4
25. a. s2 = MSE = SSE / (n- 2) = 5.30 / 3 = 1.77
s= = =:S 177 133. .
b. 2( ) 22.8ix x =
1 2
1.330.28
22.8( )b
i
ss
x x= = =
t
b
sb= = =
1
1
51
28 182
.
. .
Using ttable (3 degrees of freedom), area in tail is between .05 and .10
p-value is between .10 and .20
Actualp-value = .1664
Becausep-value > , we cannot rejectH0: 1= 0;xandydo not appear to be related.
c. MSR = SSR/1 = 5.90 /1 = 5.90
F= MSR/MSE = 5.90/1.77 = 3.33
Using Ftable (1 degree of freedom numerator and 3 denominator), p-value is greater than .10
Actualp-value = .1664
Becausep-value > , we cannot rejectH0: 1= 0;xandydo not appear to be related.
26. a. s2 = MSE = SSE / (n- 2) = 85,135.14 / 4 = 21,283.79
s= = =:S 21 28379 14589, . .
2( ) 0.74ix x =
1 2
145.89169.59
0.74( )b
i
ss
x x= = =
t b
sb= = =1
1
58108
16959343
.
..
Using ttable (4 degrees of freedom), area in tail is between .01 and .025
14 - 20
-
8/13/2019 SBE_SM14.doc
21/44
Simple Linear Regression
p-value is between .02 and .05Actualp-value = .0266
Becausep-value , we rejectH0: 1= 0
b. MSR = SSR / 1 = 249,864.86 / 1 = 249.864.86
F = MSR / MSE = 249,864.86 / 21,283.79 = 11.74
Using Ftable (1 degree of freedom numerator and 4 denominator), p-value is between .025 and .05
Actualp-value = .0266
Becausep-value , we rejectH0: 1= 0
c.
Source of Variation Sum of Squares Degrees of Freedom Mean Square F
Regression 249864.86 1 249864.86 11.74
Error 85135.14 4 21283.79
Total 335000 5
27. a. Summations needed to compute the slope andy-intercept are:
237 1654 ( )( ) 315.2 ( ) 10.1i i i i ix y x x y y x x = = = =
1 2
( )( ) 315.231.2079
10.1( )
i i
i
x x y yb
x x
= = =
0 1 165.4 (31.2079)(3.7) 19.93b y b x= = =
19.93 31.21y x= +
b. SSE =2( ) 2487.66i iy y = SST =
2( )iy y = 12,324.4
Thus, SSR = SST - SSE = 12,324.4 - 2487.66 = 9836.74
MSR = SSR/1 = 9836.74
MSE = SSE/(n- 2) = 2487.66/8 = 310.96
F = MSR / MSE = 9836.74/310.96 = 31.63
F.05 = 5.32 (1 degree of freedom numerator and 8 denominator)
Since F= 31.63 > F.05= 5.32 we rejectH0: 1= 0.
Upper support and price are related.
c. r2= SSR/SST = 9,836.74/12,324.4 = .80
The estimated regression equation provided a good fit; we should feel comfortable using theestimated regression equation to estimate the price given the upper support rating.
d. y = 19.93 + 31.21(4) = 144.77
14 - 21
-
8/13/2019 SBE_SM14.doc
22/44
Chapter 14
28. SST = 411.73 SSE = 161.37 SSR = 250.36
MSR = SSR / 1 = 250.36
MSE = SSE / (n- 2) = 161.37 / 13 = 12.413
F= MSR / MSE = 250.36 / 12.413= 20.17
Using Ftable (1 degree of freedom numerator and 13 denominator), p-value is less than .01
Actualp-value = .0006
Becausep-value , we rejectH0: 1= 0.
29. SSE = 233,333.33 SST = 5,648,333.33 SSR = 5,415,000
MSE = SSE/(n- 2) = 233,333.33/(6 - 2) = 58,333.33
MSR = SSR/1 = 5,415,000
F = MSR / MSE = 5,415,000 / 58,333.25 = 92.83
Source of Variation Sum of Squares Degrees of Freedom Mean Square F
Regression 5,415,000.00 1 5,415,000 92.83
Error 233,333.33 4 58,333.33
Total 5,648,333.33 5
Using Ftable (1 degree of freedom numerator and 4 denominator), p-value is less than .01
Actualp-value = .0006
Becausep-value , we rejectH0: 1= 0. Production volume and total cost are related.
30. Using the computations from Exercise 22,
SSE = 1,678,294 SST = 5,729,911 SSR = 4,051,617
s2= MSE = SSE/(n-2) = 1,678,294/8 = 209,786.8
209,786.8 458.02s= =
2( )ix x = 291.389
1 2
458.0226.832
291.389( )b
i
ss
x x= = =
1
1 117.9173 4.39526.832b
bt
s= = =
Using ttable (1 degree of freedom numerator and 8 denominator), area in tail is less than .005
p-value is less than .01
Actualp-value = .0024
14 - 22
-
8/13/2019 SBE_SM14.doc
23/44
Simple Linear Regression
Becausep-value , we rejectH0: 1= 0
There is a significant relationship between xandy.
31. SST = 11,373.09 SSE = 2017.37 SSR = 9355.72
MSR = SSR / 1 = 9355.72
MSE = SSE / (n- 2) = 2017.37/ 16 = 126.0856
F= MSR / MSE = 9355.72/ 126.0856 = 74.20
Using Ftable (1 degree of freedom numerator and 16 denominator), p-value is less than .01
Actualp-value = .0024
Becausep-value , we rejectH0: 1= 0.
32. a. s = 2.033
23 ( ) 10ix x x= =
'
2 2'
2
( )1 1 (4 3)2.033 1.11
5 10( )y
i
x xs s
n x x
= + = + =
b. . . . . ( ) .y x= + = + =0 2 2 6 0 2 2 6 4 10 6
; y t sy' ' 2
10.6 3.182 (1.11) = 10.6 3.53
or 7.07 to 14.13
c.
2 2'
#nd 2
( )1 1 (4 3)1 2.033 1 2.32
5 10( )i
x xs s
n x x
= + + = + + =
d. ;y t s' #nd 2
10.6 3.182 (2.32) = 10.6 7.38
or 3.22 to 17.98
33. a. s = 1.453
b. 23.8 ( ) 30.8ix x x= =
'
2 2'
2
( )1 1 (3 3.8)1.453 .068
5 30.8( )y
i
x xs s
n x x
= + = + =
14 - 23
-
8/13/2019 SBE_SM14.doc
24/44
Chapter 14
. . . . ( ) .y x= = =30 33 188 30 33 188 3 24 69
; y t sy' ' 2
24.69 3.182 (.68) = 24.69 2.16
or 22.53 to 26.85
c.
2 2'
#nd 2
( )1 1 (3 3.8)1 1.453 1 1.61
5 30.8( )i
x xs s
n x x
= + + = + + =
d. ;y t s' #nd 2
24.69 3.182 (1.61) = 24.69 5.12
or 19.57 to 29.81
34. s = 1.33
25.2 ( ) 22.8ix x x= =
'
2 2'
2
( )1 1 (3 5.2)1.33 0.85
5 22.8( )y
i
x xs s
n x x
= + = + =
. . . . ( ) .y x= + = + =0 75 0 51 0 75 051 3 2 28
; y t sy' ' 2
2.28 3.182 (.85) = 2.28 2.70
or -.40 to 4.98
2 2'
#nd 2
( )1 1 (3 5.2)1 1.33 1 1.58
5 22.8( )i
x xs s
n x x
= + + = + + =
;y t s' #nd 2
2.28
3.182 (1.58) = 2.28
5.03
or -2.27 to 7.31
35. a. s = 145.89
23.2 ( ) 0.74ix x x= =
14 - 24
-
8/13/2019 SBE_SM14.doc
25/44
Simple Linear Regression
'
2 2'
2
( )1 1 (3 3.2)145.89 68.54
6 0.74( )y
i
x xs s
n x x
= + = + =
1790.5 581.1 1790.5 581.1(3) 3533.8y x= + = + =
; y t sy' ' 2
3533.8 2.776 (68.54) = 3533.8 190.27
or $3343.53 to $3724.07
b.
2 2'
#nd 2
( )1 1 (3 3.2)1 145.89 1 161.19
6 0.74( )i
x xs s
n x x
= + + = + + =
;y t s' #nd 2
3533.8
2.776 (161.19) = 3533.8
447.46
or $3086.34 to $3981.26
36. a. 51.82 .1452 51.82 .1452(200) 80.86y x= + = + =
b. s = 3.5232
2183.4667 ( ) 11,867.73ix x x= =
'
2 2'
2
( )1 1 (200 183.4667)3.5232 1.055
15 11,867.73( )y
i
x xs s
n x x
= + = + =
; y t sy' ' 2
80.86 2.160 (1.055) = 80.86 2.279
or 78.58 to 83.14
c.
2 2'
#nd 2
( )1 1 (200 183.4667)1 3.5232 1 3.678
15 11,867.73( )i
x xs s
n x x
= + + = + + =
;y t s' #nd 2
80.86 2.160 (3.678) = 80.86 7.944
or 72.92 to 88.80
37. a. 257 ( ) 7648ix x x= =
s2= 1.88 s= 1.37
14 - 25
-
8/13/2019 SBE_SM14.doc
26/44
Chapter 14
'
2 2'
2
( )1 1 (52.5 57)1.37 0.52
7 7648( )y
i
x xs s
n x x
= + = + =
; y t sy' ' 2
13.08 2.571 (.52) = 13.08 1.34
or 11.74 to 14.42 or $11,740 to $14,420
b. sind= 1.47
13.08 2.571 (1.47) = 13.08 3.78
or 9.30 to 16.86 or $9,300 to $16,860
c. Yes, $20,400 is much larger than anticipated.
d. Any deductions exceeding the $16,860 upper limit could suggest an audit.
38. a. . . (500) $5046.y= + =1246 67 7 6 67
b. 2575 ( ) 93,750ix x x= =
s2= MSE = 58,333.33 s= 241.52
2 2'
#nd 2
( )1 1 (500 575)1 241.52 1 267.50
6 93,750( )i
x xs s
n x x
= + + = + + =
;y t s' #nd 2
5046.67 4.604 (267.50) = 5046.67 1231.57
or $3815.10 to $6278.24
c. Based on one month, $6000 is not out of line since $3815.10 to $6278.24 is the prediction interval.However, a sequence of five to seven months with consistently high costs should cause concern.
39. a. Letx= miles of track andy= weekday ridership in thousands.
Summations needed to compute the slope andy-intercept are:
2203 309 ( )( ) 1471 ( ) 838i i i i ix y x x y y x x = = = =
1 2
( )( ) 14711.7554
838( )
i i
i
x x y yb
x x
= = =
0 1 44.1429 (1.7554)(29) 6.76b y b x= = =
6.76 1.755y x= +
b. SST =3620.9 SSE = 1038.7 SSR = 2582.1
14 - 26
-
8/13/2019 SBE_SM14.doc
27/44
Simple Linear Regression
r2= SSR/SST = 2582.1/3620.9 = .713
The estimated regression equation explained 71.3% of the variability iny; a good fit.
c. s2= MSE = 1038.7/5 = 207.7
207.7 14.41s= =
'
2 2'
2
( )1 1 (30 29)14.41 5.47
7 838( )y
i
x xs s
n x x
= + = + =
6.76 1.755 6.76 1.755(30) 45.9y x= + = + =
; y t sy' ' 2
45.9 2.571(5.47) = 45.9 14.1
or 31.8 to 60
d.
2 2'
#nd 2
( )1 1 (30 29)1 14.41 1 15.41
7 838( )i
x xs s
n x x
= + + = + + =
;y t s' #nd 2
45.9 2.571(15.41) = 45.9 39.6
or 6.3 to 85.5
The prediction interval is so wide that it would not be of much value in the planning process. Alarger data set would be beneficial.
40. a. 9
b. y = 20.0 + 7.21x
c. 1.3626
d. SSE = SST - SSR = 51,984.1 - 41,587.3 = 10,396.8
MSE = 10,396.8/7 = 1,485.3
F= MSR / MSE = 41,587.3 /1,485.3 = 28.00
Using Ftable (1 degree of freedom numerator and 7 denominator), p-value is less than .01
Actualp-value = .0011
Becausep-value , we reject H0:B1= 0.
e. y = 20.0 + 7.21(50) = 380.5 or $380,500
41. a. y = 6.1092 + .8951x
14 - 27
-
8/13/2019 SBE_SM14.doc
28/44
Chapter 14
b.1
1 1 .8951 0 6.01.149b
b Bt
s
= = =
Using ttable (1 degree of freedom numerator and 8 denominator), area in tail is less than .005
p-value is less than .01
Actualp-value = .0004
Becausep-value , we reject H0:B1= 0
c. y = 6.1092 + .8951(25) = 28.49 or $28.49 per month
42 a. y = 80.0 + 50.0x
b. 30
c. F= MSR / MSE = 6828.6/82.1 = 83.17
Using Ftable (1 degree of freedom numerator and 28 denominator), p-value is less than .01
Actualp-value = .0001
Becausep-value
-
8/13/2019 SBE_SM14.doc
29/44
Simple Linear Regression
e. The 95% prediction interval is 24,285,484 to 40,667,127 .
44. a/b. The scatter diagram shows a linear relationship between the two variables.
c. The Minitab output is shown below:
The regression equation isouse = 41.& 1.33 acancy(
Predictor Coe !" Coe T P
Constant 41.490 $.333 $.&& 0.001
acancy( -1.3301 0.3&22 -3.7# 0.013
! = 3.0#923 '-!q = 74.0( '-!q)ad* = $#.#(
,naysis o ariance
!ource / !! ! P
'egression 1 13$.07 13$.07 14.2$ 0.013
'esidua "rror & 47.72 9.&4
Tota $ 1#3.79
d. Since thep-value = 0.013 is less than = .05, the relationship is significant.
e. 2r = .688. The least squares line does not provide a very good fit.
f. The 95% confidence interval is 14.53 t 20.56or NT$14,530 to NT$20,560.
g. The 95% prediction interval is 4.55 t 22.57or NT$4,550 to NT$22,570.
45. a. 270 76 ( )( ) 200 ( ) 126i i i i ix y x x y y x x = = = =
1 2( )( ) 200 1.5873
126( )i i
i
x x y ybx x
= = =
0 1 15.2 (1.5873)(14) 7.0222b y b x= = =
7.02 1.59y x= +
b. The residuals are 3.48, -2.47, -4.83, -1.6, and 5.22
c.
14 - 29
-
8/13/2019 SBE_SM14.doc
30/44
Chapter 14
*6
*4
*2
0
2
4
6
0 5 10 15 20 25
x
6e%#d
!a)%
-
8/13/2019 SBE_SM14.doc
31/44
Simple Linear Regression
29.4 1.55y x= +
b. SS/ = 1002 SS = 310.28 SS = 691.72
:S = SS ; 1 = 691.72
:S = SS ; (n* 2) = 310.28; 5 = 62.0554
F= :S ; :S = 691.72; 62.0554= 11.15
Using Ftable (1 degree of freedom numerator and 5 denominator), p-value is between .01 and .025
Actualp-value = .0206
Becausep-value = .05, e n!de tat te t a+#abe% a+e +eated.
.
*15
*10
*5
0
5
10
25 30 35 40 45 50 55 60 65
+ed#&ted @a)!e%
6e%#d!a)%
d. /e +e%#d!a 't ead% !% t ?!e%t#n te a%%!"'t#n a #nea+ +eat#n%#' beteenxandy. ent! te +eat#n%#' #% %#n##ant at te .05 ee %#n##ane, #t !d be e-t+e"edane+!% t e-t+a'ate bend te +ane te data.
14 - 31
-
8/13/2019 SBE_SM14.doc
32/44
Chapter 14
48. a. 80 4y x= +
*8
*6
*4
*2
0
2
4
6
8
0 2 4 6 8 10 12 14
x
6e%#d!a)%
b. /e a%%!"'t#n% ne+n#n te e+++ te+" a''ea+ +ea%nabe.
49. a. Letx= +et!+n n #ne%t"ent () andy= '+#e;ea+n#n% (;) +at#.
32.13 3.22y x= +
b.
*1.5
*1
*0.5
0
0.5
1
1.5
2
0 10 20 30 40 50 60
x
Standa+d#>ed
6e%#d!
a)%
. /e+e #% an !n!%!a t+end #n te +e%#d!a%. /e a%%!"'t#n% ne+n#n te e+++ te+" a''ea+?!e%t#nabe.
14 - 32
-
8/13/2019 SBE_SM14.doc
33/44
Simple Linear Regression
50. a. The MINITAB output is shown below:
The regression equation is
Y = 66.1 + 0.402 X
Predictor Coef SE Coef T p
Constant 66.10 32.06 2.06 0.094
X 0.4023 0.2276 1.77 0.137
S = 12.62 R-sq = 38.5% R-sq(adj) = 26.1%
Analysis of Variance
SOURCE DF SS MS F p
Regression 1 497.2 497.2 3.12 0.137
Residual Error 5 795.7 159.1
Total 6 1292.9
Unusual Observations
Obs. X Y Fit SEFit Residual St.Resid
1 135 145.00 120.42 4.87 24.58 2.11R
R denotes an observation with a large standardized residual.
/e %tanda+d#>ed +e%#d!a% a+e 2.11, *1.08, .14, *.38, *.78, *.04, *.41
/e #+%t b%e+at#n a''ea+% t be an !t#e+ %#ne #t a% a a+e %tanda+d#>ed +e%#d!a.
b.
2.4+
- *
STDRESID-
-
-
1.2+ -
-
-
- *
0.0+ *
-
- * *
- *
-
-1.2+ *
-
--+---------+---------+---------+---------+---------+----YHAT
110.0 115.0 120.0 125.0 130.0 135.0
The standardized residual plot indicates that the observationx = 135,y = 145 may be an outlier;note that this observation has a standardized residual of 2.11.
14 - 33
-
8/13/2019 SBE_SM14.doc
34/44
Chapter 14
c. The scatter diagram is shown below -
Y - *
-
-
135+
-
- * * -
-
120+ * *
-
-
- *
-
105+
-
- *
----+---------+---------+---------+---------+---------+--X
105 120 135 150 165 180
The scatter diagram also indicates that the observationx = 135,y = 145 may be an outlier; theimplication is that for simple linear regression an outlier can be identified by looking at the scatterdiagram.
51. a. The Minitab output is shown below:
The regression equation is
Y = 13.0 + 0.425 X
Predictor Coef SE Coef T p
Constant 13.002 2.396 5.43 0.002
X 0.4248 0.2116 2.01 0.091
S = 3.181 R-sq = 40.2% R-sq(adj) = 30.2%
Analysis of Variance
SOURCE DF SS MS F p
Regression 1 40.78 40.78 4.03 0.091
Residual Error 6 60.72 10.12
Total 7 101.50
Unusual Observations
Obs. X Y Fit Stdev.Fit Residual St.Resid
7 12.0 24.00 18.10 1.20 5.90 2.00R
8 22.0 19.00 22.35 2.78 -3.35 -2.16RX
R denotes an observation with a large standardized residual.
X denotes an observation whose X value gives it large influence.
The standardized residuals are: -1.00, -.41, .01, -.48, .25, .65, -2.00, -2.16
The last two observations in the data set appear to be outliers since the standardized residuals forthese observations are 2.00 and -2.16, respectively.
b. Using MINITAB, we obtained the following leverage values:
14 - 34
-
8/13/2019 SBE_SM14.doc
35/44
Simple Linear Regression
.28, .24, .16, .14, .13, .14, .14, .76
MINITAB identifies an observation as having high leverage if hi> 6/n; for these data, 6/n=6/8 = .75. Since the leverage for the observationx= 22,y= 19 is .76, MINITAB wouldidentify observation 8 as a high leverage point. Thus, we conclude that observation 8 is aninfluential observation.
c. 24.0+ *
-
Y -
-
-
20.0+ *
- *
- *
-
-
16.0+ *
- *
-
- *
-
12.0+ *
-
+---------+---------+---------+---------+---------+------X
0.0 5.0 10.0 15.0 20.0 25.0
The scatter diagram indicates that the observation x = 22,y = 19 is an influential observation.
52. a. The Minitab output is shown below:
The regression equation is,5ount = 4.09 + 0.19$ edia"6
Predictor Coe !" Coe T P
Constant 4.0#9 2.1$# 1.#9 0.09$
edia"6 0.19&&2 0.03$3& &.3# 0.001
! = &.044 '-!q = 7#.3( '-!q)ad* = 7&.$(
,naysis o ariance
!ource / !! ! P
'egression 1 73&.#4 73&.#4 2#.93 0.001
'esidua "rror # 203.&1 2&.44
Tota 9 939.3&
8nusua :ser;ations
:s edia"6 ,5ount it !" it 'esidua !t 'esid
1 120 3$.30 27.&& 3.30 #.7& 2.30'
' denotes an o:ser;ation
-
8/13/2019 SBE_SM14.doc
36/44
Chapter 14
observation 1 to be an outlier.
14 - 36
-
8/13/2019 SBE_SM14.doc
37/44
Simple Linear Regression
53. a. The Minitab output is shown below:
The regression equation is
Price = 1&.2 + 0.01#1 ou5e
Predictor Coe !" Coe T P
Constant 1&.193 2.004 7. 0.000
ou5e 0.01#0#$ 0.002079 #.70 0.000
! = &.297$0 '-!q = #1.7( '-!q)ad* = #0.$(
,naysis o ariance
!ource / !! ! P
'egression 1 2123.# 2123.# 7&.$# 0.000
'esidua "rror 17 477.1 2#.1
Tota 1# 2$00.9
b. /e #n +e%#d!a 't e+##e% te%e b%e+at#n%.
Fitt ed Value
Standardized
Residual
5550454035302520
2
1
0
-1
-2
14 - 37
-
8/13/2019 SBE_SM14.doc
38/44
Chapter 14
54. a. /e :#n#tab !t'!t #% %n be
The regression equation is
!aary = 707 + 0.004#2 >tCa
Predictor Coe !" Coe T P
Constant 707.0 11#.0 &.99 0.000
>tCa 0.004#1&4 0.000#07$ &.9$ 0.000
! = 379.# '-!q = $$.4( '-!q)ad* = $4.&(
,naysis o ariance
!ource / !! ! P
'egression 1 &129071 &129071 3&.&& 0.000
'esidua "rror 1# 2&9$$47 1442
Tota 19 772&71#
8nusua :ser;ations
:s >tCa !aary it !" it 'esidua !t 'esid
$ &07217 332&.0 3149.& 33#.$ 17&.& 1.02 ?
17 1209$7 11$.2 12#9.& #$.4 -1173.3 -3.17'
' denotes an o:ser;ation ed +e%#d!a 't aa#n%t te'+ed#ted a!e% #% %n be
*4
*3
*2
*1
0
1
2
3
500 1000 1500 2000 2500 3000 3500
+ed#&ted @a!e%
Standa+d#>ed
6e%#d!a)%
55. No. Regression or correlation analysis can never prove that two variables are causally related.
56. The estimate of a mean value is an estimate of the average of allyvalues associated with the samex. The estimate of an individualyvalue is an estimate of only one of theyvalues associated with aparticularx.
57. The purpose of testing whether 1 0 = is to determine whether or not there is a significant
relationship betweenxandy. However, rejecting 1 0 = does not necessarily imply a good fit. For
example, if 1 0 = is rejected and r2is low, there is a statistically significant relationship betweenx
andybut the fit is not very good.
14 - 38
-
8/13/2019 SBE_SM14.doc
39/44
Simple Linear Regression
58. a. /e :#n#tab !t'!t #% %n be
The regression equation is
Price = 9.2$ + 0.711 !hares
Predictor Coe !" Coe T P
Constant 9.2$& 1.099 #.43 0.000
!hares 0.710& 0.1474 4.#2 0.001
! = 1.419 '-!q = 74.4( '-!q)ad* = 71.2(
,naysis o ariance
!ource / !! ! P
'egression 1 4$.7#4 4$.7#4 23.22 0.001
'esidua "rror # 1$.11$ 2.01&
Tota 9 $2.900
b. S#ne tep*a!e ++e%'nd#n t = 23.22 = .001 B = .05, te +eat#n%#' #% %#n##ant.
. 2r = .744C a d #t. /e ea%t %?!a+e% #ne e-'a#ned 74.4 te a+#ab##t #n +#e.
d. 9.26 .711(6) 13.53y= + =
59. a. /e :#n#tab !t'!t #% %n be
The regression equation is
tions = - 3.#3 + 0.29$ Co55on
Predictor Coe !" Coe T P
Constant -3.#34 &.903 -0.$& 0.&29
Co55on 0.29&$7 0.02$4# 11.17 0.000
! = 11.04 '-!q = 91.9( '-!q)ad* = 91.2(
,naysis o ariance
!ource / !! ! P
'egression 1 1&20# 1&20# 124.72 0.000
'esidua "rror 11 1341 122
Tota 12 1$&&0
b. 3.83 .296(150) 40.57y= + = C a''+-#"ate 40.6 "##n %a+e% 't#n% +ant% !t%tand#n.
. 2r = .919C a e+ d #t. /e ea%t %?!a+e% #ne e-'a#ned 91.9 te a+#ab##t #n 't#n%.
60. a. The Minitab output is shown below:
The regression equation is
Horizon = 0.275 + 0.950 S&P 500
Predictor Coef SE Coef T P
Constant 0.2747 0.9004 0.31 0.768
S&P 500 0.9498 0.3569 2.66 0.029
S = 2.664 R-Sq = 47.0% R-Sq(adj) = 40.3%
Analysis of Variance
Source DF SS MS F P
14 - 39
-
8/13/2019 SBE_SM14.doc
40/44
Chapter 14
Regression 1 50.255 50.255 7.08 0.029
Residual Error 8 56.781 7.098
Total 9 107.036
b. Since thep-value = 0.029 is less than = .05, the relationship is significant.
c. r2= .470. The least squares line does not provide a very good fit.
d. Texas Instruments has higher risk with a market beta of 1.25.
61. a.
40
50
60
70
80
90
100
35 45 55 65 75 85
L$ /e"'e+at! +e
=#2(/e"'e+at!+e
b. It a''ea+% tat te+e #% a '%#t#e #nea+ +eat#n%#' beteen te t a+#abe%.
. /e :#n#tab !t'!t #% %n be
The regression equation is
igh = 23.9 + 0.#9# @o