M3 Part II.
M3 Part II.
x (years) 6 12 14 6 9 13 15 9
y (Stock Units) 300 400 560 250 290 650 630 520
8 3930 - 12 400
r
8 104 - (12) 2 8 200000 - (400) 2
r 0.8464
Spearman’s Rank Correlation
• Used when quantitative measures of certain factors
e.g. leadership ability, efficiency, honesty,
Intelligence etc. can not be fixed
• Individuals in the groups are arranged in order and
ranked
• If n is the number of paired ranks then the
Spearman’s Rank Correlation Coefficient is given by
6d 2
R 1-
n(n 2 - 1)
• Where d denotes the difference between the paired
ranks
• Three possibilities arise when we find the Rank
Correlation Coefficient from a given set of
observations
6 20
R 1- 2
0 . 88
10(10 - 1)
Actual Ranks are Not Given
An economist wanted to find out if there was any relationship
between the unemployment rate in a country and its inflation
rate. Data gathered from 7 countries for the year 2004 are
given below:
Country Unemployment Rate (%) Inflation Rate (%)
A 4.0 3.2
B 8.5 8.2
C 5.5 9.4
D 0.8 5.1
E 7.3 10.1
F 5.8 7.8
G 2.1 4.7
UR % RANK IR % RANK d d²
A 4 3 3.2 1 2 4
B 8.5 7 8.2 5 2 4
C 5.5 4 9.4 6 -2 4
D 0.8 1 5.1 3 -2 4
E 7.3 6 10.1 7 -1 1
F 5.8 5 7.8 4 1 1
G 2.1 2 4.7 2 0 0
Sd² = 18
Actual Ranks are Not Given
6 18
R 1- 2
0 . 678
7(7 - 1)
(m
3
m ) (m
3
m2)
6 d 2 1 1
2
..........
12 12
R 1-
n(n 2 - 1)
(2 3 2) (3 3 3)
6 81.5
12 12 504
R 1- 2
1 0
8(8 - 1) 504
HOMEWORK
At a debate competition, ten contestants are ranked by three
Judges as follows. Use the rank correlation coefficient to discuss
which pair of judges have the nearest approach.
A B C D E F G H I J
J1 10 5 6 1 8 9 7 2 4 3
J2 8 6 3 7 4 1 9 10 5 2
J3 5 7 2 3 10 9 8 1 6 4
Method of Concurrent Deviations
• A simple method which gives us a quick but crude idea of
correlation between two variables.
Steps:
1. Add three columns dx, dy and dx.dy to the table of
values.
2. Find out the direction of change of each value of x
variable as compared with the previous value and write in
‘dx’ column . If increasing denote by ‘+’, if decreasing,
denote by ‘ – ‘, if constant denote by 0.
3. Repeat the procedure for y variable and write the change
in dy column.
4. Multiply dx and dy and count the number of ‘+” signs in
dxdy column. This number is denoted by c.
5. Apply the following formula:
(2c m)
rc
m
Where,
rc Coefficient of Correlation
and m n - 1
Clearly, c = 0 and m = 7 – 1 = 6
HOMEWORK
2 2 2
J1 J2 J3 d(1.2) d(2.3) d(3.1)
A 10 8 5 4 9 25
B 5 6 7 1 1 4
C 6 3 2 9 1 16
D 1 7 3 36 16 4
E 8 4 10 16 36 4
F 9 1 9 64 64 0
G 7 9 8 4 1 1
H 2 10 1 64 81 1
I 4 5 6 1 1 4
J 3 2 4 1 4 1
200 214 60
R 1.2 0 . 21 R 2.3 0 . 30 R 3.1 0 . 64