In: Statistics and Probability
Use table below to analyze correlation and to develop regression equations describing the relationships between person’s height and weight. Make separate regression lines for men and women.
gender |
height |
weight |
0 |
72 |
155 |
0 |
67 |
145 |
0 |
65 |
125 |
1 |
67 |
120 |
1 |
63 |
105 |
1 |
54 |
120 |
1 |
66 |
125 |
1 |
64 |
125 |
0 |
72 |
160 |
0 |
66 |
133 |
0 |
66 |
175 |
0 |
71 |
205 |
0 |
71 |
175 |
1 |
57 |
82 |
1 |
66 |
125 |
1 |
67 |
133 |
0 |
70 |
175 |
0 |
73 |
163 |
1 |
68 |
133 |
0 |
72 |
180 |
1 |
65 |
107 |
1 |
70 |
170 |
1 |
64 |
110 |
1 |
64 |
140 |
1 |
63 |
110 |
1 |
60 |
110 |
1 |
69 |
125 |
1 |
65 |
120 |
1 |
67 |
180 |
0 |
67 |
120 |
0 |
68 |
140 |
0 |
65 |
130 |
1 |
62 |
122 |
1 |
66 |
114 |
1 |
65 |
115 |
1 |
63 |
125 |
0 |
73 |
195 |
0 |
69 |
135 |
0 |
70 |
145 |
0 |
72 |
170 |
0 |
73 |
172 |
0 |
69 |
168 |
0 |
68 |
155 |
0 |
71 |
185 |
0 |
71 |
175 |
0 |
68 |
158 |
0 |
69 |
185 |
0 |
67 |
146 |
0 |
66 |
135 |
0 |
67 |
150 |
0 |
72 |
160 |
0 |
68 |
155 |
0 |
75 |
230 |
0 |
68 |
149 |
0 |
73 |
240 |
0 |
72 |
170 |
0 |
72 |
198 |
0 |
72 |
163 |
0 |
72 |
230 |
0 |
74 |
170 |
0 |
68 |
151 |
0 |
73 |
220 |
0 |
68 |
145 |
0 |
70 |
130 |
0 |
72 |
160 |
0 |
70 |
210 |
0 |
67 |
145 |
0 |
67 |
185 |
0 |
71 |
237 |
0 |
72 |
205 |
0 |
73 |
147 |
0 |
68 |
170 |
0 |
72 |
181 |
0 |
68 |
150 |
0 |
67 |
150 |
0 |
70 |
200 |
0 |
71 |
175 |
0 |
70 |
155 |
0 |
67 |
167 |
1 |
58 |
100 |
1 |
56 |
100 |
Soln
i)
Assuming for Males, Gender = 0
height (Y) |
weight (X) |
XY |
Y2 |
X2 |
|
72 |
155 |
11160 |
5184 |
24025 |
|
67 |
145 |
9715 |
4489 |
21025 |
|
65 |
125 |
8125 |
4225 |
15625 |
|
72 |
160 |
11520 |
5184 |
25600 |
|
66 |
133 |
8778 |
4356 |
17689 |
|
66 |
175 |
11550 |
4356 |
30625 |
|
71 |
205 |
14555 |
5041 |
42025 |
|
71 |
175 |
12425 |
5041 |
30625 |
|
70 |
175 |
12250 |
4900 |
30625 |
|
73 |
163 |
11899 |
5329 |
26569 |
|
72 |
180 |
12960 |
5184 |
32400 |
|
67 |
120 |
8040 |
4489 |
14400 |
|
68 |
140 |
9520 |
4624 |
19600 |
|
65 |
130 |
8450 |
4225 |
16900 |
|
73 |
195 |
14235 |
5329 |
38025 |
|
69 |
135 |
9315 |
4761 |
18225 |
|
70 |
145 |
10150 |
4900 |
21025 |
|
72 |
170 |
12240 |
5184 |
28900 |
|
73 |
172 |
12556 |
5329 |
29584 |
|
69 |
168 |
11592 |
4761 |
28224 |
|
68 |
155 |
10540 |
4624 |
24025 |
|
71 |
185 |
13135 |
5041 |
34225 |
|
71 |
175 |
12425 |
5041 |
30625 |
|
68 |
158 |
10744 |
4624 |
24964 |
|
69 |
185 |
12765 |
4761 |
34225 |
|
67 |
146 |
9782 |
4489 |
21316 |
|
66 |
135 |
8910 |
4356 |
18225 |
|
67 |
150 |
10050 |
4489 |
22500 |
|
72 |
160 |
11520 |
5184 |
25600 |
|
68 |
155 |
10540 |
4624 |
24025 |
|
75 |
230 |
17250 |
5625 |
52900 |
|
68 |
149 |
10132 |
4624 |
22201 |
|
73 |
240 |
17520 |
5329 |
57600 |
|
72 |
170 |
12240 |
5184 |
28900 |
|
72 |
198 |
14256 |
5184 |
39204 |
|
72 |
163 |
11736 |
5184 |
26569 |
|
72 |
230 |
16560 |
5184 |
52900 |
|
74 |
170 |
12580 |
5476 |
28900 |
|
68 |
151 |
10268 |
4624 |
22801 |
|
73 |
220 |
16060 |
5329 |
48400 |
|
68 |
145 |
9860 |
4624 |
21025 |
|
70 |
130 |
9100 |
4900 |
16900 |
|
72 |
160 |
11520 |
5184 |
25600 |
|
70 |
210 |
14700 |
4900 |
44100 |
|
67 |
145 |
9715 |
4489 |
21025 |
|
67 |
185 |
12395 |
4489 |
34225 |
|
71 |
237 |
16827 |
5041 |
56169 |
|
72 |
205 |
14760 |
5184 |
42025 |
|
73 |
147 |
10731 |
5329 |
21609 |
|
68 |
170 |
11560 |
4624 |
28900 |
|
72 |
181 |
13032 |
5184 |
32761 |
|
68 |
150 |
10200 |
4624 |
22500 |
|
67 |
150 |
10050 |
4489 |
22500 |
|
70 |
200 |
14000 |
4900 |
40000 |
|
71 |
175 |
12425 |
5041 |
30625 |
|
70 |
155 |
10850 |
4900 |
24025 |
|
67 |
167 |
11189 |
4489 |
27889 |
|
Total |
3980 |
9603 |
672962 |
278258 |
1663699 |
Using the above values and formula, we get:
r = 0.603
The magnitude of r indicates moderate correlation between height and weight for males and the positive sign indicates direct relationship ie as height increases, weight also increases and vice versa
Let the regression equation be: Y = a + bX
Where
Slope(b) = {n*∑XY - ∑X *∑Y}/{n*∑X2 – (∑X)2 } = 0.05
and a = ∑Y/n – b*∑X/n = 60.87
Hence, Height = 60.87 + 0.05 Weight
ii)
Females:
gender |
height (Y) |
weight (X) |
XY |
Y2 |
X2 |
1 |
67 |
120 |
8040 |
4489 |
14400 |
1 |
63 |
105 |
6615 |
3969 |
11025 |
1 |
54 |
120 |
6480 |
2916 |
14400 |
1 |
66 |
125 |
8250 |
4356 |
15625 |
1 |
64 |
125 |
8000 |
4096 |
15625 |
1 |
57 |
82 |
4674 |
3249 |
6724 |
1 |
66 |
125 |
8250 |
4356 |
15625 |
1 |
67 |
133 |
8911 |
4489 |
17689 |
1 |
68 |
133 |
9044 |
4624 |
17689 |
1 |
65 |
107 |
6955 |
4225 |
11449 |
1 |
70 |
170 |
11900 |
4900 |
28900 |
1 |
64 |
110 |
7040 |
4096 |
12100 |
1 |
64 |
140 |
8960 |
4096 |
19600 |
1 |
63 |
110 |
6930 |
3969 |
12100 |
1 |
60 |
110 |
6600 |
3600 |
12100 |
1 |
69 |
125 |
8625 |
4761 |
15625 |
1 |
65 |
120 |
7800 |
4225 |
14400 |
1 |
67 |
180 |
12060 |
4489 |
32400 |
1 |
62 |
122 |
7564 |
3844 |
14884 |
1 |
66 |
114 |
7524 |
4356 |
12996 |
1 |
65 |
115 |
7475 |
4225 |
13225 |
1 |
63 |
125 |
7875 |
3969 |
15625 |
1 |
58 |
100 |
5800 |
3364 |
10000 |
1 |
56 |
100 |
5600 |
3136 |
10000 |
Total |
1529 |
2916 |
186972 |
97799 |
364206 |
Using the above values and formula, we get:
r = 0.610
The magnitude of r indicates moderate correlation between height and weight for females and the positive sign indicates direct relationship ie as height increases, weight also increases and vice versa
Let the regression equation be: Y = a + bX
Where
Slope(b) = {n*∑XY - ∑X *∑Y}/{n*∑X2 – (∑X)2 } = 0.12
and a = ∑Y/n – b*∑X/n = 49.02
Hence, Height = 49.02 + 0.12 Weight