STA 4210 – Fall 2003

advertisement
STA 4210 – Fall 2003
Exam 1 – Take Home Portion
x1
x2
x3
x4
x5
x6
x7
x8
x9
x10
x11
x12
0
0
10
10
20
20
30
30
40
40
50
50
name
y1
y2
y3
y4
y5
y6
y7
y8
y9
y10
y11
y12
ABALO
BEKKER
CARROLL
CASTILLO
CLARK
CROOKSHANKS
DOBSON
DOLINSKAYA
DONAHUE
DUNN
GIDEON
HENDERSON
HU
JOHNSON
KLEIN
LATHAM
MINOR
O'BRIEN
PARRIS
PIMENTEL
PROBY
RAHN
RAMIREZ
SCHEBESTA
SCHERER
TERKEURST
WILLS
YEN
ZADENSKY
ZDON
9
9
14
15
6
11
14
13
12
7
11
16
18
11
6
11
11
10
14
9
10
6
11
5
12
7
11
9
8
10
7
9
9
10
8
6
13
8
11
13
12
14
7
4
10
17
14
11
11
14
12
12
15
6
12
8
14
9
12
9
28
28
34
31
31
26
28
29
26
32
32
35
32
30
22
27
28
31
23
35
33
33
29
34
34
24
34
31
29
24
32
27
33
30
28
33
31
33
36
34
32
32
25
31
31
31
31
31
33
27
41
28
31
30
35
26
32
26
33
29
51
50
49
52
50
51
48
48
46
50
47
54
54
48
45
49
56
52
46
52
46
48
49
49
50
50
52
54
54
50
52
47
48
51
53
50
44
53
47
53
51
57
49
49
51
50
54
52
46
52
52
55
56
47
51
49
53
49
55
55
74
66
63
68
68
66
76
69
68
68
74
66
69
67
67
70
69
69
72
67
71
67
69
69
65
74
69
71
73
67
68
70
73
75
70
71
75
72
69
76
76
68
66
70
73
67
63
72
75
73
72
79
68
71
71
74
68
66
66
72
87
83
84
88
94
84
93
90
92
87
89
92
87
91
88
91
95
87
87
93
90
86
93
87
92
91
90
94
90
85
94
91
96
87
82
87
90
90
91
88
90
88
88
86
93
86
86
91
95
89
85
91
86
91
87
87
91
88
94
90
110
110
114
114
109
107
116
115
110
110
107
114
113
114
120
110
105
110
112
102
108
109
110
107
106
109
104
113
108
112
109
116
108
115
111
114
113
114
110
99
114
115
108
110
104
109
107
102
116
107
104
113
112
110
113
111
108
107
113
113
Each student has n=12 observations, at the X levels given at the top of the page. Neatly complete the
following parts, based on the simple linear regression model with independent and normally distributed
errors of constant variance. This is due at beginning of Exam 1 on Friday, Sept. 26. No late exams will
be accepted.
1) Obtain the least squares fitted equation. Give all fitted values and residuals.
2) Give the error sum of squares and the mean square error.
3) Give a 95% confidence interval for 1.
4) Give a 95% confidence interval for 0
5) Give a 95% confidence interval for E{Y|X=35}
6) Give a 95% prediction interval for a new observation when X=35
7) Give the Analysis of Variance
8) Use the F-test to test whether 1=0 (=0.05)
9) Plot the residuals versus the fitted values.
10) Give a normal probability plot of the residuals.
11) Use the correlation test to test whether the residuals are normally distributed.
12) Use the modified-Levene test for constant error variance.
Download