Scribd
Upload
67 views23 pages

Correlation Coefficient Calculation

This document discusses correlation and regression analysis. It defines correlation as a measure of the relationship between two variables on a scale of -1 to 1. Regression finds the best fitting linear relationship between two correlated variables. The regression line of y on x takes the form y = ax + b, while the regression line of x on y is x = ay + b. Formulas are provided to calculate the coefficients a and b of the regression lines based on the variables' means, standard deviations, and correlation coefficient.

Uploaded by

S LIKITH
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
67 views23 pages

Correlation Coefficient Calculation

This document discusses correlation and regression analysis. It defines correlation as a measure of the relationship between two variables on a scale of -1 to 1. Regression finds the best fitting linear relationship between two correlated variables. The regression line of y on x takes the form y = ax + b, while the regression line of x on y is x = ay + b. Formulas are provided to calculate the coefficients a and b of the regression lines based on the variables' means, standard deviations, and correlation coefficient.

Uploaded by

S LIKITH
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

MODULE – 04

Statistical Methods
Correlation and Regression

This topic deals with data concerning independent observations.

Examples:

1. Marks of individuals in two subjects


2. Height and weight of individuals
3. Amount involved in advertising a product and sale of product etc.,
We discuss the aspect of inter – relation between the independent variables.

 Correlation and Correlation Coefficient


The numerical measure of correlation between two variables x and y is known as

Pearson’s coefficient of correlation usually denoted by r and is defined as follows.

∑𝑛
1 (𝑥− 𝑥̅ )(𝑦− 𝑦
̅)
r= … (1)
𝑛𝜎𝑥 𝜎𝑦

This can be put in the alternative form as follows.

If X = x -𝑥̅ , Y = y-𝑦̅ we can write

Thus (1) becomes

∑ 𝑋𝑌
r=
√∑ 𝑋 2 ∑ 𝑌 2

Property :

The coeeficient of correlation numerically does not exceed unity.

ie -1 ≤ r ≤ +1.

[Link] MATHS,GMIT Page 1


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

Note : If r = ± 1 we say that x and y are perfectly correlated and if r = 0 we say that x and

y are non correlated.

 Alternative formula for the Correlation Coefficient r

𝜎𝑥2 +𝜎𝑦2 −𝜎𝑥−𝑦


2
r=
2𝜎𝑥 𝜎𝑦

Problems:

1. Calculate the karl – pearson co – efficient for the following ages of husband and
wife’s.
Roll No. 1 2 3 4 5 6 7 8 9 10
Husband’s 36 23 27 28 28 29 30 31 33 35
age (x)
Wife’s age 29 18 20 22 27 21 29 27 29 28
(y)

∑𝑋 300 ∑𝑌 250
Solu: 𝑥̅ = = = 30 𝑦̅ = = = 25 Here n = 10
10 10 10 10

x y X= x - 𝑥̅ Y = y -𝑦̅ XY X2 Y2
36 29 -7 -7 49 49 49
23 18 -3 -5 15 9 25
27 20 -2 -3 6 4 9
28 22 -2 2 -4 4 4
28 27 -1 -4 4 1 16
29 21 0 4 0 0 16
30 29 1 2 2 1 4
31 27 3 4 12 9 16
33 29 5 3 15 25 9
35 28 6 4 24 36 16
∑𝑥 ∑𝑦 ∑ 𝑋𝑌 ∑ 𝑋 2 ∑ 𝑌 2
= 300 = 250 = 123 = 138 = 164

[Link] MATHS,GMIT Page 2


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

We have

∑ 𝑋𝑌
r=
√∑ 𝑋 2 ∑ 𝑌 2
123
r= = 0.817
√(138)(164)

It is a positive correlation.

2. Obtain the correlation of the following data:

x 10 14 18 22 26 30
y 18 12 24 6 30 36

∑ 𝑋𝑌
Solu. We have r= where X = x - 𝑥̅ , Y = y -𝑦̅
√∑ 𝑋 2 ∑ 𝑌 2

Here n = 6
∑𝑋 120 ∑𝑌 126
𝑥̅ = = = 20 𝑦̅ = = = 21
10 6 10 6

x y X= x -𝑥̅ Y = y-𝑦̅ XY X2 Y2
10 18 -10 -3 30 100 9

14 18 -6 -9 53 36 81

18 18 -2 3 -6 4 9

22 18 2 -15 -30 4 225

26 18 6 9 54 36 81

30 18 10 15 150 100 225

∑𝑥 ∑𝑦 ∑ 𝑋𝑌 ∑ 𝑋 2 ∑ 𝑌 2
= 120 = 126 = 252 = 280 = 6304

[Link] MATHS,GMIT Page 3


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

252
We have r= = 0.6
√(280)(630)

It is a positive correlation.

3. Calculate Karl pearson co-efficient of correlation b/w the marks obtained by 8 students in
mathematics and statistics :

Statistics 8 10 15 17 20 23 24 25
Mathematics 25 30 32 35 37 40 42 45

Solu: Let statistics = x , Mathematics = y

∑ 𝑋𝑌
We have r= where X = x - 𝑥̅ , Y = y -𝑦̅
√∑ 𝑋 2 ∑ 𝑌 2

Here n = 8
∑𝑋 142 ∑𝑌 286
𝑥̅ = = = 17.75 𝑦̅ = = = 35.75
𝑛 8 𝑛 8

x y X= x -𝑥̅ Y = y-𝑦̅ XY X2 Y2
8 25 -9.75 -10.75 104.81 95.06 115.56
10 30 -7.75 -5.75 44.56 60.06 33.06
15 32 -2.75 -3.75 10.31 7.56 14.06
17 35 -0.75 -0.75 0.56 0.56 0.56
20 37 2.25 1.25 2.81 5.06 1.56
23 40 5.25 4.25 22.31 27.56 18.06
24 42 6.25 6.25 39.06 39.06 39.06
25 45 7.25 9.25 67.06 56.56 85.56

∑𝑥 ∑𝑦 ∑ 𝑋𝑌 ∑ 𝑋2 ∑ 𝑌2
= 17.75 = 35.75 = 291.48 = 291.480 = 307.48

291.48
We have r= = 0.97
√(291.48)(307.48)

It is a positive correlation.

[Link] MATHS,GMIT Page 4


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

Regression
Regression is an estimation of one independent variable in terms of the other. If x & y are
correlated, the best fitting straight line in the least square sense gives reasonably a good relation
between x & y.

The best fitting straight line of the form y = ax + b (x being the independent variable) is called
the regression line of y on x & x = ay + b (y being the independent variable) is called the
regression line of x on y.

Formulas for line of regression

Let y = ax + b be the equation of the regression line of y on x for a given set of n values (x,y).
𝜎𝑦
Then y- 𝑦̅ = 𝑟 (x-𝑥̅ )………….(1)
𝜎𝑥

This is the regression of y on x.

Similarly,
𝜎𝑥
x- 𝑥̅ = 𝑟 (y-𝑦̅)………..(2)
𝜎𝑦

This is the regression line of x on y.


𝜎𝑦 𝜎
The coefficient of x in (1) & the coefficient of y in (2) respectively given by r and r 𝜎𝑥 are
𝜎𝑥 𝑦

known as the regression coefficients.

Their product is equal to 𝑟 2 .

Thus we can conclude that r is the geometric mean (GM) of the regression coefficients since the
GM of two numbers a, b is √𝑎𝑏. That is

r = ±√(𝑐𝑜𝑒𝑓𝑓. 𝑜𝑓 𝑥 )(𝑐𝑜𝑒𝑓𝑓. 𝑜𝑓 𝑦)

The sign of r will be positive or negative according as the regression coefficients are positive or
negative.

Note: The lines of regression (1) & (2) are also of the form
∑ 𝑋𝑌 ∑ 𝑋𝑌
Y= ∑ 𝑋2
(X) and X = ∑ 𝑌 2 (Y)

Where X= x- 𝑥̅ & Y= y- 𝑦̅.

[Link] MATHS,GMIT Page 5


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

This form will be useful to find out the coefficient of correction by first obtaining the lines of
regression as we have deduced that

r = ±√(𝑐𝑜𝑒𝑓𝑓. 𝑜𝑓 𝑥 )(𝑐𝑜𝑒𝑓𝑓. 𝑜𝑓 𝑦)

Problems:

1. Compute the coefficient of correlation & the equation of the lines of regression for the
following data.

x 1 2 3 4 5 6 7
y 9 8 10 12 11 13 14

∑ 𝑋𝑌
Solu: We have coefficient of correlation r =
√∑𝑋 2 ∑𝑌 2

Where Y = y- 𝑦̅ X= x- 𝑥̅ Here n = 7
∑𝑥 ∑𝑦
𝑥̅ = = 28/7 = 4 𝑦̅ = = 77/7 = 11
𝑛 𝑛

x y X= x- Y= y- 𝑦̅ XY 𝑋2 𝑌2
𝑥̅
1 9 -3 -2 6 9 4
2 8 -2 -3 6 4 9
3 10 -1 -1 1 1 1
4 12 0 1 0 0 1
5 11 1 0 0 1 0
6 13 2 2 4 4 4
7 14 3 3 9 9 9
2
∑ 𝑋 =28 ∑ 𝑌 2 =28
∑𝑥 ∑𝑦 ∑ 𝑋𝑌
= 28 = 77 = 26

Line regression y on x Line regression x on y


∑ 𝑋𝑌 ∑ 𝑋𝑌
Y = ∑ 𝑋2 . X X= ∑ 𝑌2
.Y

26 26
y- 𝑦̅ = (x- 𝑥̅ ) x-4 = 28 (y-11)
28

y-11 = 0.928 (x-4) x-4 = 0.928y -10.208

y = 0.92x + 7.29 x = 0.928y-6.208


[Link] MATHS,GMIT Page 6
COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

2. Obtain the lines of regression and hence find the co-efficient of correlation for the
following data

x 1 3 4 2 5 8 9 10 13 15
y 8 6 10 8 12 16 16 10 32 32
Solution:
∑𝑥 ∑𝑦
𝑥̅ = = 70/10 = 7 𝑦̅ = = 150/10 =15
𝑛 𝑛

x y X= x- 𝑥̅ Y= y- 𝑦̅ xy 𝑋2 𝑌2
1 8 -6 -7 42 36 49
3 6 -4 -9 35 16 81
4 10 -3 -5 15 9 25
2 8 -5 -7 35 25 49
5 12 -2 -3 6 4 9
8 16 1 1 1 1 1
9 16 2 1 2 4 1
10 10 3 -5 -15 9 25
13 32 6 17 102 36 289
15 32 8 17 136 64 289
∑𝑥 ∑𝑦 ∑ 𝑋𝑌 ∑ 𝑋2 ∑ 𝑌2
= 10 = 150 = 360 = 204 = 818

Line of regression y on x line of regression x on y


∑ 𝑋𝑌 ∑ 𝑋𝑌
Y = ∑ 𝑋2 . X X= ∑ 𝑌2
.Y

360
y- 𝑦̅ = 360/204( x- 𝑥̅ ) x- 7 = 818 (y-15)

y-15 = 1.764 (x-7) x- 7 = 0.44y + 6.6

y-15 = 1.764x-12.348 x = 0.44y + 0.4

y = 1.764x-12.348+15

y = 1.764x + 2.652

We have

Co-efficient of correlation r = ±√(𝑐𝑜𝑒𝑓𝑓. 𝑜𝑓 𝑥 )(𝑐𝑜𝑒𝑓𝑓. 𝑜𝑓 𝑦)

= 0.88

[Link] MATHS,GMIT Page 7


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

3. Compute the coefficient of correlation & the equation of the lines of regression for the
following data.

x 10 14 18 22 26 30
y 18 12 24 6 30 36

∑𝑥 120 ∑𝑦 126
Solu: 𝑥̅ = = = 20 𝑦̅ = = = 21
𝑛 6 𝑛 6

x y X= x- 𝑥̅ Y= y- 𝑦̅ XY 𝑋2 𝑌2
10 18 -10 -3 30 100 9
14 12 -6 -9 54 36 81
18 24 -2 3 -6 4 9
22 6 2 -15 -30 4 225
26 30 6 9 54 36 81
30 36 10 15 150 100 225
∑ 𝑋 2 =280 ∑ 𝑌 2 =630
∑𝑥 ∑𝑦 ∑ 𝑋𝑌
= 120 = 126 = 252

∑ 𝑋𝑌
We have, Coefficient of correlation r = = 0.6
√∑𝑋 2 ∑𝑌 2

Line of regression y on x line of regression x on y


∑ 𝑋𝑌 ∑ 𝑋𝑌
Y = ∑ 𝑋2 . X X= ∑ 𝑌2
.Y

252 252
y- 𝑦̅ = 280 ( x- 𝑥̅ ) ( x- 𝑥̅ ) = (y- 𝑦̅ )
630

y – 21 = 0.9 ( x – 20) x – 20 = 0.4 ( y – 21)

y = 0.9 x – 18+21 x = 0.4y -8.4 + 20

y = 0.9 x + 3 x = 0.4 y +11.6

[Link] MATHS,GMIT Page 8


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

4. If ϴ is the acute angle between the lines of regression, then show that
𝜎𝑥 𝜎𝑦 1− 𝑟 2
tan ϴ = ( ).Explain the significance when r = 0 & r = ± 1.
𝜎𝑥 2 + 𝜎𝑦 2 𝑟

Solu: W. K. T If is ϴ acute, the angle between the lines y = m1 x + c1 and y = m2 x +c2 is given by
𝑚2 − 𝑚1
tan ϴ =
1+ 𝑚1 𝑚2

We have the lines of regression ,


𝜎𝑦
y - 𝑦̅ = r ( x - 𝑥̅ ) …. (1)
𝜎𝑥

𝜎𝑥
and ( x - 𝑥̅ ) = r (y - 𝑦̅)
𝜎𝑦

𝜎𝑦
We write the second of the equation as y - 𝑦̅ = ( x - 𝑥̅ ) ….(2)
𝑟𝜎𝑥

Slope of (1) and (2) are respectively given by


𝜎𝑦 𝜎𝑦
m1 = r and m2 =
𝜎𝑥 𝑟𝜎𝑥

Substituting these in the formula for tan ϴ we have,


𝜎𝑦 𝜎𝑦
r 𝜎 − 𝑟𝜎
𝑥 𝑥
tan ϴ = 𝜎𝑦 𝜎𝑦
1+ r
𝜎𝑥 𝑟𝜎𝑥

𝜎𝑥 𝜎𝑦 1− 𝑟 2
tan ϴ = ( )
𝜎𝑥 2 + 𝜎𝑦 2 𝑟

If r = ± 1 , tan ϴ = 0 → ϴ = 0,which implies that the two regression lines coincide


𝜋
and hence the variables are perfectely correlated. Also if r = 0 , tan ϴ = ∞ or ϴ = .
2
This implies that the lines are perpendicular and hence the variables are uncorrelated.

[Link] MATHS,GMIT Page 9


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

5. In a partially destroyed record,only the lines of regression of y on x and x on y are


available as 4x – 5y + 33 = 0 and 20x – 9y = 107 respectively. Calculate 𝑥 , 𝑦 and
the coefficient of correlation between x and y.

Solu. W .K .T regression lines passes through 𝑥̅ and 𝑦̅

4𝑥̅ -5𝑦̅ = -33 and 20𝑥̅ - 9𝑦̅ = 107

By solving we get 𝑥̅ = 13 𝑦̅ = 17

We shall now rewrite the equation of the regression lines to find the regression

coefficients.

5y = 4x + 33 or y = 0.8 x + 6.6 ….(1)

20x = 9y + 107 or x = 0.45y + 4.35 …(2)

r = ± √(𝑐𝑜𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑥 )(𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑦)

= ± √(0.8)(0.45)

r = 0.6

It is a positive correlation.

[Link] MATHS,GMIT Page 10


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

Rank Correlation and an expression for the rank correlation coefficient.

The coefficient of correlation in respect of the ranks of some two characteristics of an


individual or an observation is called Rank Correlation Coefficient usually denoted by ρ

We now proceed to derive an expression for ρ in the following form.

6 ∑(𝑥−𝑦)2 6 ∑ 𝑑2
ρ =1- or 1 -
𝑛 (𝑛2 − 1) 𝑛 (𝑛2 − 1)

Note:

(1) If the ranking of x ,y are entirely in the same order like for example, x :1 1,2,3,4,5 ; y :
1,2,3,4,5 then ∑ 𝑑 2 = ∑(𝑥 − 𝑦 )2 = 0. This will give us 𝜌 = ± 1 and is called perfect direct
correlation.
If the ranking of x and y are entirely in the opposite order like for example, x : 1,2,3,4,5
y : 5,4,3,2,1 then ∑ 𝑑 2 = [Link] will give us ρ = -1 and is called perfect inverse
correlation.

Problems:

1. Ten competitors in a beauty contest are ranked by two judges in the following
order. Compute the coefficient of correlation
I 1 6 5 3 10 2 4 9 7 8
II 6 4 9 8 1 2 3 10 5 7

6 ∑ 𝑑2
Solu : We have ρ =1-
𝑛 (𝑛2 − 1)

For the given data, n = 10 and

∑ 𝑑2 = (1 − 6)2 + (6 − 4)2 +(5 − 9)2 + (3 − 8)2 +(10 − 1)2 +(2 − 2)2 +(4 − 3)2 + (9 − 10)2

+ (7 − 5)2 +(8 − 7)2

= 25 + 4+ 16 + 25 + 81 + 0 +1 +1 + 4 + 1 =158
6 (158)
Hence 𝜌 = 1 – 10(102 − 1) = 0.042

[Link] MATHS,GMIT Page 11


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

2. Ten students got the following percentage of marks in two subjects x and y.
Compute their rank correlation coefficient.

Marks in 78 36 98 25 75 82 90 62 65 39
x
Marks in 84 51 91 60 68 62 86 58 53 47
y

Solu : We prepare the table consisting of the given data along with the ranks assigned

according to their order of the [Link] the subject x, 98 will be awarded rank

1, 90 as rank 2 and so on.

Marks in x Rank(x) Marks in y Rank(y) d = (x-y) 𝑑 2 = (𝑥 − 𝑦)2


78 4 84 3 1 1
36 9 51 9 0 0
98 1 91 1 0 0
25 10 60 6 4 16
75 5 68 4 1 1
82 3 62 5 -2 4
90 2 86 2 0 0
62 7 58 7 0 0
65 6 53 8 -2 4
39 8 47 10 -2 4
∑ 𝑑 2 = 30

6 ∑ 𝑑2
We have ρ =1- and n = 10 for the given data.
𝑛 (𝑛2 − 1)
6 (30)
= 1 - 10(102 − 1)

= 0.82

[Link] MATHS,GMIT Page 12


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

3. Ten competitors in music contest are ranked by 3 judges A, B,C in the following order.

Use the rank correlation coefficient to decide which pair of judges have the nearest

. approach to common taste of music

A 1 6 5 10 3 2 4 9 7 8
B 3 5 8 4 7 10 2 1 6 9
C 6 4 9 8 1 2 3 10 5 7

Solu : We shall compute 𝜌𝐴𝐵 , 𝜌𝐵𝐶 , 𝜌𝐶𝐴 with the help of the following table where d is

the difference in ranks.


2 2 2
A B C 𝑑𝐴𝐵 𝑑𝐵𝐶 𝑑𝐶𝐴
1 3 6 4 9 25
6 5 4 1 1 4
5 8 9 9 1 16
10 4 8 36 16 4
3 7 1 16 36 4
2 10 2 64 64 0
4 2 3 4 1 1
9 1 10 64 81 1
7 6 5 1 1 4
8 9 7 1 4 1
2 2 2
∑ 𝑑𝐴𝐵 ∑ 𝑑𝐵𝐶 ∑ 𝑑𝐶𝐴
= 200 = 214 = 60

6 ∑ 𝑑2
We have ρ = 1 - 𝑛 (𝑛2− 1) and n = 10 for the given data.

6 (200)
Now, 𝜌𝐴𝐵 = 1 - 10(102 − 1) = - 0.21

6 (214)
𝜌𝐵𝐶 = 1 - 10(102 − 1) = - 0.297

6 (60)
𝜌𝐶𝐴 = 1 - = + 0.636
10(102 − 1)

It may be observed that 𝜌𝐴𝐵 and 𝜌𝐵𝐶 are negative which means their tastes

(A &B; B &C) are opposite. But 𝜌𝐶𝐴 is positive and is nearer to 1.(perfect

correlation)

[Link] MATHS,GMIT Page 13


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 14


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 15


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 16


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 17


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 18


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 19


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 20


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 21


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 22


COMPLEX ANALYSIS,PROBABILITY AND STATISTICAL METHODS (18MAT41)

[Link] MATHS,GMIT Page 23

Common questions

Powered by AI

The Karl Pearson coefficient of correlation method can be used to determine the correlation between two variables. This method involves calculating the Pearson correlation coefficient (r) through the formula r = ∑XY / √(∑X² * ∑Y²), where X and Y are deviations from their respective means . Regression lines can also be derived using these deviations to establish the best fitting line equations, expressed as y = ax + b or x = ay + b, where a and b are constants derived from the dataset .

To determine which pair of judges has the closest alignment of tastes, calculate the rank correlation coefficient (ρ) for each judge pair using their rank data. The formula used is ρ = 1 - 6∑d² / n(n² - 1), where d represents the differences in ranks between the judges for each contestant. The pair with the ρ value closest to 1 has the most similar taste, indicating high agreement in ranking the contestants .

To calculate the coefficient of correlation between marks in two subjects, we use the formula r = ΣXY / √(ΣX² * ΣY²), where X and Y are deviations from the mean. First, calculate the means of the two datasets. Then, determine the deviations, compute the product of deviations (XY), and squares of deviations (X² and Y²). Substitute these sums into the formula to find r. A high positive r value suggests strong positive academic performance correlation, indicating students scoring high in one subject tend to score high in the other .

Given partially destroyed records with regression equations, you can determine the mean values (𝑥̅ and 𝑦̅) by noting that the regression lines always pass through the mean points of the variables. By solving the system of equations formed by the regression lines, which can be rewritten in the form y = ax + c and x = by + d, you substitute the equations into each other to find the required mean values .

The lines of regression are used to calculate the coefficient of correlation (r) by first finding the regression coefficients from the regression line equations. Once the regression lines are established, the coefficient of correlation is determined by the product of the regression coefficients, expressed as r = ±√((coeff. of x)(coeff. of y)). This result implies the strength and direction of a linear relationship between the variables. A value close to 1 or -1 indicates a strong linear relationship, while a value near 0 suggests a weak or no linear relationship .

The regression coefficients, which are rσy/σx for y on x and rσx/σy for x on y, are related to the correlation coefficient (r) such that their product is r². This indicates that r is the geometric mean of these coefficients. This relationship is significant because it confirms that the correlation coefficient is derived from the consistent proportionality in the relationships described by the regression lines. This means a high correlation (close to ±1) suggests that the regression coefficients are also significant, indicating a strong linear relationship between the variables .

In a regression context, the coefficient of correlation stands as a statistical measure that reflects the strength and direction of a linear relationship between two quantitative variables, such as student grades in different subjects. A high positive coefficient implies that as one grade increases, the other tends to increase as well, suggesting a potential link between subject performances. Conversely, a low or zero correlation indicates that the grades in one subject offer no prediction or are independent of the other subject's grades. This insight is crucial for understanding academic profiles and forming effective educational strategies .

To compute the rank correlation coefficient, denoted as ρ, for two sets of ranks, we use the formula ρ = 1 - 6∑d² / n(n² - 1), where d is the difference between the pairs of ranks and n is the number of observations. This coefficient provides insight into the degree to which a monotonic relationship exists between two sets of data. A ρ value of 1 indicates a perfect agreement in the rank ordering, while -1 indicates a perfect inverse relationship .

When the correlation coefficient (r) is zero, the regression lines are perpendicular, implying that the variables are uncorrelated, indicating no linear relationship between them . Conversely, when r is ±1, the regression lines coincide, suggesting that there is a perfect linear relationship between the variables. This means that one variable can be perfectly predicted from the other, either in a direct or inverse manner .

The assertion that the geometric mean of the regression coefficients equals the correlation coefficient (r) is based on the formula r = ±√((coeff. of x)(coeff. of y)). This expression shows that the correlation coefficient is derived from the consistent multiplicative relationship between the magnitudes of the regression coefficients for y on x and x on y. Since these coefficients indicate the slope of the regression lines, the geometric mean linking them to r emphasizes their joint contribution to measuring linear dependency between variables .

You might also like