F.Y.B.

COM
Sem-II

Mathematical & Statistical Techniques-II

Prof. Anil Khadse NKTT College, TThane 1

 Syllabus Sem-II

Mathematics
Module-I : Derivatives & its Applications Marks:20
Module-II : -Interest
-Annuity Marks:20
Statistics
Module-III : -Correlation
- Regression Marks:20
Module-IV : -Index Number
-Time series Marks:20
Module-V : Probability Distribution Marks:20
Prof. Anil Khadse NKTT College, TThane 2
Prof. Anil Khadse NKTT College, TThane 3
Correlation
Correlation means Finding the relationship between two
quantitative variables without being able to infer causal
relationships

Correlation is a statistical technique used to determine

the degree to which two variables are related

e.g Demand & Supply,

Income & Expenditure
Age and Height of Children
Study Hours & Score
Prof. Anil Khadse NKTT College, TThane 4
•Are two variables related?
• Does one increase as the other increases?
• e. g. skills and income
• Does one decrease as the other increases?
• e. g. health problems and nutrition
•How can we get a numerical measure of the
degree of relationship?

Prof. Anil Khadse NKTT College, TThane 5

Types of Correlation

Positive Correlation: Negative Correlation:

If two series move in same If two series move in opposite
direction that is one increases direction that is one increases
other also increases or both other decreases or vice a versa
decreases, there is positive ,there is negative correlation
correlation between the between the variables.
variables. Example:
Example: Price & Demand
Income & Expenditure
Study hrs & score
Age of Husband & wife Prof. Anil Khadse NKTT College, TThane 6
Methods to determine Correlation

 Scatter Diagram

 Karl Pearson’s Correlation coefficient

 Spearman’s Rank Correlation Coefficient

Prof. Anil Khadse NKTT College, TThane 7

Scatter Diagram
Scatter Diagram gives us the idea about existence of the
relation between variables.

In scatter diagram one of the variable is consider on X-axis and

other on Y-axis.

The points are plotted on graph and from the direction of the
movement of the points we can conclude on the relationship

Prof. Anil Khadse NKTT College, TThane 8

Positive Correlation Negative Correlation

No Correlation
Perfect Positive Perfect Negative

Prof. Anil Khadse NKTT College, TThane 9

Positive Correlation
Prof. Anil Khadse NKTT College, TThane 10
 Negative relationship

Reliability

Age of Car
Prof. Anil Khadse NKTT College, TThane 11
Karl Pearson’s Coefficient of correlation (r)/Product Moment

If 𝑥1, 𝑦1 ,𝑥2, 𝑦2 −−− − 𝑥𝑛, 𝑦𝑛 𝑎𝑟𝑒 𝑛 𝑝𝑎𝑖𝑟𝑠 𝑜𝑓

𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠 𝑜𝑓 𝑡𝑤𝑜 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒 𝑋 𝑎𝑛𝑑 𝑌, 𝐾𝑎𝑟𝑙 𝑃𝑒𝑎𝑟𝑠𝑜𝑛 𝐶𝑜𝑒𝑓𝑓.
𝑜𝑓 𝑐𝑜𝑟𝑟𝑒𝑙𝑎𝑡𝑖𝑜𝑛 𝑠 𝑑𝑒𝑛𝑜𝑡𝑒𝑑 𝑏𝑦 𝑟 𝑎𝑛𝑑 𝑑𝑒𝑓𝑖𝑛𝑒𝑑 𝑎𝑠

(𝑥−𝑥 )(𝑦−𝑦 ) - - - - - - - - - (i)

𝑟= 𝑛.𝜎𝑥.𝜎𝑦

Where, 𝑥 = 𝑀𝑒𝑎𝑛 𝑜𝑓 𝑥 𝑣𝑎𝑙𝑢𝑒𝑠

𝑦 = 𝑀𝑒𝑎𝑛 𝑜𝑓 𝑦 𝑣𝑎𝑙𝑢𝑒𝑠
𝜎𝑥 = s.d of x values
𝜎𝑦 = s,d of y values
Prof. Anil Khadse NKTT College, TThane 12
 (𝑥−𝑥 )(𝑦−𝑦 ) - - - - - - - - - (ii)
𝑟=
𝑥−𝑥 2 . 𝑦−𝑦 2

𝐶𝑜𝑣(𝑥,𝑦) - - - - - - - - - (iii)
𝑟= 𝑛.𝜎𝑥.𝜎𝑦

After substituting the values of mean and s.d

𝑥𝑦− 𝑥 𝑦/𝑛
𝑟= - - - - - - - - - (iv)
2 ( 𝑥)2 ( 𝑦)2
2
𝑥− . 𝑦−
𝑛 𝑛
Prof. Anil Khadse NKTT College, TThane 13
 Interpretation of r :

 The value of r ranges between ( -1) and ( +1)

 The value of r denotes the strength of the association as
illustrated by the following diagram.

strong intermediate weak weak intermediate strong

-1 -0.75 -0.25 0 0.25 0.75 1

indirect Direct
perfect perfect
correlation correlation
no relation
Prof. Anil Khadse NKTT College, TThane 14
1. If 0 < r < 1, then positive correlation
2. If - 1 < r < 0, then Negative correlation

3. If r = +1, then Perfect positive correlation

4. If r = -1, then Perfect negative correlation

5. If r = 0, then No correlation

Prof. Anil Khadse NKTT College, TThane 15

Example # 1
Calculate Karl Pearson’s coefficient of correlation for the following data

X 12 10 8 13 7
Y 15 20 25 18 22

x y 𝑥−𝑥 𝑦−𝑦 𝑥−𝑥 2 (𝑦 − 𝑦)2 (𝑥 − 𝑥 )( 𝑦 − 𝑦 )

12 15 2 -5 4 25 -10
10 20 0 0 0 0 0
8 25 -2 5 4 25 -10
13 18 3 -2 9 4 -6
7 22 -3 2 9 4 -6
50 100 26 58 -32
Prof. Anil Khadse NKTT College, TThane 16
 (𝑥−𝑥 )(𝑦−𝑦 )
𝑟=
𝑥−𝑥 2 . 𝑦−𝑦 2

−32
𝑟=
26 . 58
−32
𝑟= 5.099× .7.6157

−32
𝑟= 38.8324

𝑟 = −0.8240
There is –ve correlation
Prof. Anil Khadse NKTT College, TThane 17
Example # 2
Find Karl Pearson’s coefficient of correlation for the following data
serial Age Weight
No (years) (Kg)
1 7 12
2 6 8
3 8 12
4 5 10
5 6 11
6 9 13

Prof. Anil Khadse NKTT College, TThane 18

Sr.No x y xy X2 Y2
1 7 12 84 49 144
2 6 8 48 36 64
3 8 12 96 64 144
4 5 10 50 25 100
5 6 11 66 36 121
6 9 13 117 81 169
Total Ʃx= Ʃy= Ʃx2= Ʃy2=
Ʃxy=
41 66 461 291 742

Prof. Anil Khadse NKTT College, TThane 19

 𝑥𝑦− 𝑥 𝑦/𝑛
𝑟=
2 ( 𝑥)2 ( 𝑦)2
2
𝑥− . 𝑦−
𝑛 𝑛
41 66
461 
r 6
 (41)2   (66)2 
291  .742  
 6  6 

461−451
𝑟=
291−280.1666 . 742−726
10
𝑟= 10 𝑟 = 13.1656
10.8334 . 16

𝑟= 10 𝑟 =0.7595
3.2914×4
Prof. Anil Khadse NKTT College, TThane 20
Example #3 : Calculate Coefficient of correlation from the following
information
𝑛 = 12 , Ʃ𝑥 = 35, Ʃ𝑦 = 60, Ʃ𝑥2=148 , Ʃ𝑦2 = 450, Ʃ𝑥𝑦 = 105

𝑥𝑦− 𝑥 𝑦/𝑛
𝑟=
2 ( 𝑥)2 2 ( 𝑦)2
𝑥− . 𝑦−
𝑛 𝑛

105−35×60/12
𝑟=
35 2 60 2
148− . 450−
12 12
105−175
𝑟=
1225 3600
148− . 450−
12 12
Prof. Anil Khadse NKTT College, TThane 21
 −70
𝑟=
148− 102.0833 . 450− 300

−70
𝑟=
45.9187 . 150
−70
𝑟= 6.7763 .12.2474

−70
𝑟 = 82.9923

𝑟 = −0.84

Prof. Anil Khadse NKTT College, TThane 22

Example #4: Calculate Coefficient of correlation from the following results
𝑛 = 8 , 𝑥 − 𝑥 2 = 84, 𝑦 − 𝑦 2 =158 , (𝑥 − 𝑥 )(𝑦 − 𝑦 ) = 111

(𝑥−𝑥 )(𝑦−𝑦 )
𝑟=
𝑥−𝑥 2 . 𝑦−𝑦 2

111
𝑟=
84 . 158
111
𝑟= 9.1651 .12.5698
111 𝑟 = 0.9635
𝑟= 115..2035
Prof. Anil Khadse NKTT College, TThane 23
Example #5: Calculate Coefficient of correlation from the following results
𝑛 = 8 , 𝑠. 𝑑 𝑜𝑓 𝑥 = 3.86, 𝑠, 𝑑 𝑜𝑓 𝑦 = 6.57 , (𝑥 − 𝑥 )(𝑦 − 𝑦 ) = 192

(𝑥−𝑥 )(𝑦−𝑦 )
𝑟= 𝑛.𝜎𝑥.𝜎𝑦
192
𝑟= 8×3.86×6.57
192
𝑟= 202.8816

𝑟 = 0.9463 = 0.95

Prof. Anil Khadse NKTT College, TThane 24

 Exercise # 1
Calculate Karl Pearson’s coefficient of correlation for the following data

X 14 8 10 11 9 13 5
Ans: 0.9231
Y 14 9 11 13 11 12 4

Excercise#2 : Calculate Coefficient of correlation from the following

information
𝑛 = 25 , Ʃ𝑥 = 75, Ʃ𝑦 = 100, Ʃ𝑥2=250 , Ʃ𝑦2 = 500, Ʃ𝑥𝑦 = 325

Example #3: Calculate Coefficient of correlation from the following results

𝑛 = 8 , 𝑥 − 𝑥 2 = 100, 𝑦 − 𝑦 2 =195 , (𝑥 − 𝑥 )(𝑦 − 𝑦 ) = 85

Prof. Anil Khadse NKTT College, TThane 25

Spearman’s Rank Correlation Coefficient (R)
It is applicable for quantitative as well as qualitative data
The given values of the variable are assigned the rank in order,
and these rank provides data to calculate correlation. It is
denoted by R and defined as
6 𝑑2
𝑅 =1− 𝑛(𝑛2−1)

Where, 𝑑 = 𝑅1 − 𝑅2: 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑟𝑎𝑛𝑘𝑠

𝑅1 = 𝑅𝑎𝑛𝑘 𝑓𝑜𝑟 𝑓𝑖𝑟𝑠𝑡 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒 𝑜𝑟 𝑥
𝑅2 = 𝑅𝑎𝑛𝑘 𝑓𝑜𝑟 𝑆𝑒𝑐𝑜𝑛𝑑 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒 or y
𝑛 = 𝑁𝑜. 𝑜𝑓 𝑝𝑎𝑖𝑟𝑠 𝑜𝑓 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛
Prof. Anil Khadse NKTT College, TThane 26
Example #1: Calculate Rank correlation coefficient from the following
Rank by Judge -A 5 2 1 4 3
Rank by Judge-B 4 2 3 5 1

R1 R2 d=R1-R2 d2
5 4 1 1
2 2 0 0
1 3 -2 4
4 5 -1 1
3 1 2 4
Ʃd2= 10
Prof. Anil Khadse NKTT College, TThane 27
n=5
6 𝑑2
𝑅 =1− 𝑛(𝑛2−1)
6×10
𝑅 =1− 5(52−1)
60
𝑅 =1− 5(25−1)
60
𝑅 =1− 5(24)
60
𝑅 =1− 120

𝑅 = 1 −0.5 = 0.5
Prof. Anil Khadse NKTT College, TThane 28
Example #2: Find Rank correlation coefficient from the following data
Demand 15 22 20 30 25
Supply 18 25 26 28 20
Sol: Let R1 denotes rank for demand and R2 denotes rank for supply
x y R1 R2 d d2
15 18 5 5 0 0
22 25 3 3 0 0
20 26 4 2 2 4
30 28 1 1 0 0
25 20 2 4 -2 4
Ʃd2= 8
Prof. Anil Khadse NKTT College, TThane 29
n=5
6 𝑑2
𝑅 =1− 𝑛(𝑛2−1)
6×8
𝑅 =1− 5(52−1)
48
𝑅 =1− 5(25−1)
48
𝑅 =1− 5(24)
48
𝑅 =1− 120

𝑅 = 1 −0.4 = 0.6
Prof. Anil Khadse NKTT College, TThane 30
Example# 3
The coefficient of rank correlation between marks in two subjects
obtained by a group of students is 0.8. If the sum of squares of the
differences in ranks is 33. Find the number of students in the group.
Given: Rank correlation R =0.8
𝑑2= 33, n =?
198
6 𝑑2 −0.2 = −𝑛(𝑛2−1)
𝑅 = 1 − 𝑛(𝑛2−1)
𝑛(𝑛 2 − 1) =198/0.2
6×33
0.8 = 1 − 𝑛(𝑛2−1)
𝑛(𝑛2 − 1) =198/0.2=990
6×33
0.8 − 1 = −𝑛(𝑛2−1) 𝑛(𝑛2 − 1) =10(102-1)
𝑛 =10
Prof. Anil Khadse NKTT College, TThane 31
Rank correlation for repeated values :

If the values are repeated then the common rank is assigned by

considering average of the ranks to all repeated values.

If the values are repeated then the correction factor is added to

𝑑2 to find rank correlation
𝑚(𝑚2−1)
𝐶. 𝐹 = 12

𝐶. 𝐹 = 1
12
[(m1(m12-1)+m2(m22-1)+------]
6( 𝑑2+𝐶.𝐹)
𝑅 =1− 𝑛 𝑛 2
−1
Prof. Anil Khadse NKTT College, TThane 32
Example #2: Find Rank correlation coefficient from the following data
X 25 20 20 18 32 35
y 58 55 48 62 55 40
Sol: Let R1 denotes rank for demand and R2 denotes rank for supply
x y R1 R2 d d2
20 repeated twice
25 58 3 2 1 1
Rank = 4+5/2=4.5
20 55 4.5 3.5 1 1 m1 = 2
20 48 4.5 5 -0.5 0.25 55 repeated twice
18 62 6 1 5 25 Rank = 3+4/2=3.5
32 55 2 3.5 -1.5 2.25 m2 = 2
35 40 1 6 -5 25
Prof. Anil Khadse NKTT College, TThane Ʃd2= 54.58 33
𝐶. 𝐹 = 1
12
[(m1(m12-1)+m2(m22-1)]
1
𝐶. 𝐹 = 12 [(2(22-1)+2(22-1)]
1
𝐶. 𝐹 = 12 [(2(4-1)+2(4-1)]
1
𝐶. 𝐹 = 12 [6+6)]=12/12=1

6( 𝑑2+𝐶.𝐹)
𝑅 =1− 𝑛 𝑛2−1
6(54.5+1)
𝑅 = 1 − 6 62−1
333
𝑅 = 1 − 6 36−1
Prof. Anil Khadse NKTT College, TThane 34
 333
𝑅 =1− 6 35
333
𝑅 =1− 210

𝑅 = 1 − 1.5857

𝑅 = −0.5857

Prof. Anil Khadse NKTT College, TThane 35

Prof. Anil Khadse NKTT College, TThane 36
Regression
Regression method is use to predict / Estimate one
variable (dependent) when the value of independent
variable is known
Tells you how values in y change as a function of changes
in values of x

1.Businessman wants to know effect of increase in

advertising on sales.
2.To find the effect of change in demand pattern of some
commodities on price
Prof. Anil Khadse NKTT College, TThane 37
 Types of Regression

Regression Equation of y on x

Regression Equation of x on y

Prof. Anil Khadse NKTT College, TThane 38

1. Regression equation of y on x
It is used to estimate the value of dependant variable y
when the value of independent variable x is given
Regression equation of y on x is given by
𝑦 = 𝑎 + 𝑏𝑥
𝑦 − 𝑦 = 𝑏𝑦𝑥(𝑥 − 𝑥 )
𝑦 = 𝑏𝑦𝑥 𝑥 − 𝑥 + 𝑦
𝑥 = 𝑚𝑒𝑎𝑛 𝑜𝑓 𝑋, 𝑦 = 𝑚𝑒𝑎𝑛 𝑜𝑓 𝑌
𝑥. 𝑦
𝑥𝑦 − 𝑛
𝑏𝑦𝑥 = 𝑟𝑒𝑔𝑟𝑒𝑠𝑠𝑖𝑜𝑛 𝑐𝑜𝑒𝑓𝑓 𝑜𝑓 𝑦 𝑜𝑛 𝑥 𝑏𝑦𝑥 = ( 𝑥)2
σ𝑦 𝑥2− 𝑛
𝑏𝑦𝑥 = 𝑟.
σ𝑥 Prof. Anil Khadse NKTT College, TThane 39
2. Regression equation of x on y
It is used to estimate the value of dependant variable x
when the value of independent variable y is given

Regression equation of x on y is given by

𝑥 = 𝑎 + 𝑏y
𝑥 − 𝑥 = 𝑏𝑥𝑦(𝑦 − 𝑦)
𝑥 = 𝑏𝑥𝑦 𝑦 − 𝑦 + 𝑥
𝑥. 𝑦
𝑥𝑦 − 𝑛
𝑏𝑥𝑦 = 𝑟𝑒𝑔𝑟𝑒𝑠𝑠𝑖𝑜𝑛 𝑐𝑜𝑒𝑓𝑓 𝑜𝑓 𝑥 𝑜𝑛 y 𝑏𝑥𝑦 = ( 𝑦)2
σ𝑥 𝑦2− 𝑛
𝑏𝑥𝑦 = 𝑟.
σ𝑦 Prof. Anil Khadse NKTT College, TThane 40
 Linear Equations
Y
ŷY =bX
a +bX
a
Change
b = Slope in Y
Change in X
a = Y-intercept
X
Prof. Anil Khadse NKTT College, TThane 41
Example #1 Find two regression equations from the following data

X 8 7 10 9 5 6
y 11 8 12 13 8 10

x y x2 y2 xy
8 11 64 121 88
7 8 49 64 56
10 12 100 144 120
9 13 81 169 117
5 8 25 64 40
6 10 36 100 60
Ʃx = Ʃy = Ʃx2 Ʃy2 = Ʃxy =
45 62 355 662 481
Prof. Anil Khadse NKTT College, TThane 42
n=6 𝑥 45 𝑦 62
𝑥= 𝑛
= = 7.5 𝑦= 𝑛
= = 10.33
6 6

𝑥. 𝑦
𝑥𝑦 − 𝑛
𝑏𝑦𝑥 = ( 𝑥)2
𝑥2− 𝑛
481 − 45 × 62/6
𝑏𝑦𝑥 = 45 2
355 − 6
481 − 465
𝑏𝑦𝑥 =
355 − 337.5
Prof. Anil Khadse NKTT College, TThane 43
 16
𝑏𝑦𝑥 = = 0.91
17.5
𝑥. 𝑦
𝑥𝑦 − 𝑛
𝑏𝑥𝑦 = ( 𝑦)2
𝑦2− 𝑛
16
𝑏𝑥𝑦 = 62 2
662 − 6
16 16
𝑏𝑥𝑦 = 𝑏𝑥𝑦 = = 0.75
662 − 640.666 21.34
Prof. Anil Khadse NKTT College, TThane 44
Regression equation of y on x is given by

𝑦 = 𝑏𝑦𝑥 𝑥 − 𝑥 + 𝑦

𝑦 = 0.91 𝑥 − 7.5 + 10.33

𝑦 = 0.91𝑥 − 0.91 × 7.5 + 10.33

𝑦 = 0.91𝑥 − 6.825 + 10.33

𝑦 = 0.91𝑥 + 3.505

Prof. Anil Khadse NKTT College, TThane 45

Regression equation of x on y is given by
𝑥 = 𝑏𝑥𝑦 𝑦 − 𝑦 + 𝑥

𝑥 = 0.75 𝑦 − 10.33 + 7.5

𝑥 = 0.75𝑦 − 0.75 × 10.33 + 7.5

𝑥 = 0.75𝑦 − 7.7475 + 7.5

𝑥 = 0.75𝑦 − 0.2475

Prof. Anil Khadse NKTT College, TThane 46

Example # 2
For bivariate distribution, Mean of x = 65, Mean of y = 53
s.d of x = 4.7 s.d of y = 5.2, Correlation Coeff = 0.78
Find two regression equations and estimate
i) The most probable value of y when x = 63
ii) The most probable value of x when y = 50
Given: 𝑥 = 65 , 𝑦 = 53, σ𝑥 = 4.7, σ𝑦 = 5.2, 𝑟 = 0.78
σ𝑦 5.2
𝑏𝑦𝑥 = 𝑟. = 0.78. = 0.86
σ𝑥 4.7
σ𝑥 4.7
𝑏𝑥𝑦 = 𝑟. = 0.78. = 0.705 = 0.71
σ𝑦 5.2
Prof. Anil Khadse NKTT College, TThane 47
Regression equation of y on x is given by
𝑦 = 𝑏𝑦𝑥 𝑥 − 𝑥 + 𝑦
𝑦 = 0.86 𝑥 − 65 + 53
𝑦 = 0.86𝑥 − 0.86 × 65 + 53
𝑦 = 0.86𝑥 − 55.9 + 53
𝑦 = 0.86𝑥 − 2.9
When x=63

𝑦 = 0.86 × 63 − 2.9 𝑦 = 51.28

𝑦 = 54.18 − 2.9
Prof. Anil Khadse NKTT College, TThane 48
Regression equation of x on y is given by
𝑥 = 𝑏𝑥𝑦 𝑦 − 𝑦 + 𝑥
𝑥 = 0.71 𝑦 − 53 + 65
= 0.71𝑦 − 0.71 × 53 + 65
= 0.71𝑦 − 37.63 + 65
𝑥 = 0.71𝑦 + 27.37
When y=50

𝑥 = 0.71 × 50 + 27.37 𝑥 = 62.87

𝑥 = 35.5 + 27.37
Prof. Anil Khadse NKTT College, TThane 49
Example # 3
For bivariate distribution, Mean of x = 43, Mean of y = 37
Regression coeff. of y on x = 0.59
Regression Coeff of x on y = 0.72
Find two regression equations and estimate
i) Likely value of y when x = 40
ii) Likely value of x when y = 35
Given: 𝑥 = 43 , 𝑦 = 37, 𝑏𝑦𝑥 = 0.59, 𝑏𝑥𝑦 = 0.72,
Regression equation of y on x is given by
𝑦 = 𝑏𝑦𝑥 𝑥 − 𝑥 + 𝑦

Prof. Anil Khadse NKTT College, TThane 50

 𝑦 = 0.59 𝑥 − 43 + 37
𝑦 = 0.59𝑥 − 0.59 × 43 + 37

𝑦 = 0.59𝑥 − 25.37 + 37

𝑦 = 0.59𝑥 + 11.63

When x=40

𝑦 = 0.59 × 40 + 11.63

𝑦 = 23.6 + 11.63 = 35.23

Prof. Anil Khadse NKTT College, TThane 51

Regression equation of x on y is given by
𝑥 = 𝑏𝑥𝑦 𝑦 − 𝑦 + 𝑥
𝑥 = 0.72 𝑦 − 37 + 43
𝑥 = 0.72𝑦 − 0.72 × 37 + 43
𝑥 = 0.72𝑦 − 26.64 + 43
𝑥 = 0.72𝑦 + 16.36
When y= 35
𝑥 = 0.72 × 35 + 16.36
𝑥 = 41.56
Prof. Anil Khadse NKTT College, TThane 52
Example # 4
Given the following data, find two regression equations. Also
estimate y if x=60 and x if y = 37
x Y
Mean 65 39
s.d 4.3 1.2

Correlation Coefficient = 0.75

Given: 𝑥 = 65 , 𝑦 = 39, σ𝑥 = 4.3, σ𝑦 = 1.2, 𝑟 = 0.75
σ𝑦 1.2
𝑏𝑦𝑥 = 𝑟. 𝑏𝑦𝑥 = 0.75 × 𝑏𝑦𝑥 = 0.21
σ𝑥 4.3
Prof. Anil Khadse NKTT College, TThane 53
 σ𝑥 4.3
𝑏𝑥𝑦 = 𝑟. 𝑏𝑥𝑦 = 0.75 × = 2.69
σ𝑦 1.2
Regression equation of y on x is given by
𝑦 = 0.21 𝑥 − 65 + 39
𝑦 = 0.21𝑥 − 13.65 + 39
𝑦 = 0.21𝑥 + 25.35
𝑦 = 0.21 × 60 + 25.35 Put x = 60

𝑦 = 37.95
Prof. Anil Khadse NKTT College, TThane 54
Regression equation of x on y is given by
𝑥 = 𝑏𝑥𝑦 𝑦 − 𝑦 + 𝑥
𝑥 = 2.69 𝑦 − 39 + 65
𝑥 = 2.69𝑦 − 104.91 + 65
𝑥 = 2.69𝑦 − 39.91
Put y = 37
𝑥 = 2.69 × 37 − 39.91
𝑥 = 99.53 − 39.91
𝑥 = 59.62
Prof. Anil Khadse NKTT College, TThane 55
Properties of regression lines

 The two regression lines coincide if there is perfect +ve or

perfect – ve correlation between the variables.

 The two regression line are perpendicular to each other if

there is no correlation between the variables.
 The point 𝑥 , 𝑦 𝑠𝑎𝑡𝑖𝑠𝑓𝑖𝑒𝑠 𝑏𝑜𝑡ℎ 𝑡ℎ𝑒 𝑟𝑒𝑔𝑟𝑒𝑠𝑠𝑖𝑜𝑛 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛𝑠
 Relation between correlation and regression coefficients is

𝑟 = ± 𝑏𝑦𝑥. 𝑏𝑥𝑦
Prof. Anil Khadse NKTT College, TThane 56
Sign of corr coeff. r depends on the sign of regression coefficients.
i) r is positive if both the regression coefficients are positive
ii) r is negative if both the regression coefficients are negative

From the given regression equations, regression coefficients

can be obtained as
From reg. equation of y on x
𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑥
𝑏𝑦𝑥 = −𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑦

From reg. equation of x on y

𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑦
𝑏𝑥𝑦 = −𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑥
Prof. Anil Khadse NKTT College, TThane 57
Example # 5
The regression equation of y on x is 𝑥 + 3𝑦 − 88 = 0
and that of regression equation of x on y is 2𝑥 + 𝑦 − 71 = 0.
Find i) Mean values of x and y
ii) Coefficient of correlation
SOL: To find mean values , solve two regression equations
𝑥 + 3𝑦 − 88 = 0 −−−− −(i) ×2
2𝑥 + 𝑦 − 71 = 0 −−−− −(ii)
2𝑥 + 6𝑦 − 176 = 0 −−−− −(i)
2𝑥 + 𝑦 − 71 = 0 −−−−− − ii
- - + .
5y = 105 y = 21
Prof. Anil Khadse NKTT College, TThane 58
 Put y = 21 in equation (i)
𝑥 + 3𝑦 − 88 = 0
𝑥 + 3 × 21 − 88 = 0
𝑥 + 63 − 88 = 0
𝑥 =25
𝑥 = 25 , 𝑦 = 21
ii) To find r
The regression equation of y on x is 𝑥 + 3𝑦 − 88 = 0

𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑥
𝑏𝑦𝑥 = −𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑦
1
𝑏𝑦𝑥 = −3

Prof. Anil Khadse NKTT College, TThane 59

Regression equation of x on y is 2𝑥 + 𝑦 − 71 = 0.
𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑦 1
𝑏𝑥𝑦 = −𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑥 = −2

𝑟 = ± 𝑏𝑦𝑥. 𝑏𝑥𝑦

1 1 1
𝑟 =± − ×− =−
3 2 6
𝑟 = − 0.1666
𝑟 = −0.4082
Prof. Anil Khadse NKTT College, TThane 60
Example # 6
Given two regression equations 2𝑥 + 3𝑦 = 5 and 𝑥 + 𝑦 = 2.
Find i) Mean values of x and y
ii) Coefficient of correlation
SOL: To find mean values , solve two regression equations
2𝑥 + 3𝑦 = 5 −−−−−− −(𝑖)
𝑥+𝑦 =2 −−−−−− −(𝑖𝑖) × 2
2𝑥 + 3𝑦 = 5
2𝑥 + 2𝑦 = 4 𝑠𝑢𝑏𝑡𝑟𝑎𝑐𝑡 𝑖𝑖 𝑓𝑟𝑜𝑚 𝑖
- - -
𝑦=1
Put y = 1 in equation ii 𝑥+1=2 ⇒𝑥 = 1
𝑥 =1&𝑦 =1 Prof. Anil Khadse NKTT College, TThane 61
ii) To find r
Let regression equation of y on x is 2𝑥 + 3𝑦 = 5
𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑥 2
𝑏𝑦𝑥 = −𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑦 𝑏𝑦𝑥 = −3
Regression equation of x on y is 𝑥 + 𝑦 = 1.
1
𝑏𝑥𝑦 = −𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑦
𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑥
𝑏𝑦𝑥 = −1 = −1

𝑟 = ± 𝑏𝑦𝑥. 𝑏𝑥𝑦 𝑟 = ± −2/3 × −1

𝑟 = − 0.6666
𝑟 = −0.8165
Prof. Anil Khadse NKTT College, TThane 62
Exercise#1 Find two regression equations from the following data and
Estimate y when x=16 and x when y 18
X 3 4 6 10 12 13
y 12 11 15 16 19 17
Example # 2
For bivariate distribution, Mean of x = 25, Mean of y = 152
s.d of x = 1.8 s.d of y = 5.7, Correlation Coeff = 0.82
Find two regression equations and estimate y if x = 23 and x if y = 145
Example # 3
Given the following data, find two regression equations. Also estimate age of
wife if age of husband is 25.
Husband Wife
Average age 27 years 23 years
s.d 3 years 2 years
Correlation Coefficient = 0.75
Prof. Anil Khadse NKTT College, TThane 63
Example # 6
Given two regression equations 3𝑥 − 𝑦 − 25 = 0 and 2𝑥 − 3𝑦 + 30
= 0.
Find i) Mean values of x and y
ii) Coefficient of correlation
iii) s.d of x if s.d of y is 2

σ𝑥
𝑏𝑥𝑦 = 𝑟.
σ𝑦
σ𝑥
1/3 = 0.4714.
2
2/3 = 0.4714. σ𝑥 1.41 = σ𝑥
Prof. Anil Khadse NKTT College, TThane 64