Unit 6

Regression Analysis
six not I In no
“Regression refers to the statistical technique of modeling the relationship between

variables”.

“A regression model is a mathematical equation that describes the relationship between

two or more variables”.

inkensen
There are two types of regression analysis: y sep

nsent.F.T
1) Simple regression analysis: The link between two variables is modeled in

linear regression. The dependent variable is 201

5 w.jp
independent variable is represented by (X).
01
represented by (Y), whereas the

2) Multiple regression analysis: We create a model that shows how one

at least 3variable if
dependent variable and two or more independent variables are related.

IIII
We'll use a straight-line relationship model to show how X and Y are related.

1
 110 1 at
Regression Equation

mean
x̅
samplesize
x̅
o

Types of regressions:

simple 2variables TEETH regression line equation

E a bx
multiple at least 3variables 2 ormore independent
dependent regression line Equation

gyykatb.X.tbzfi.tk
Regression Equation

y intercept OR
constant term

sin as
Dependent t.si
2
value Independent value

I
Slope coefficient

am 6

IIIO.us
Significant effect:
Type of effect:

t
reisefeaIpossitiveeffet
Negativeeffect
obvaueis
b value is
p.FMwe 1.96 Thereis effect
Interpret of b:
ii
Model valid: Indicates that an (increase or decrease) in
If the question does not give us value of (a) it will be (X Independent) by one unit will (increase
a=0.05 or decrease) (b value) by (Y dependent)

Significance f a build
significance f a not vails
 64 1 Is

independent

Y dependent

n 10
x̅ 53

x̅ 67
EI

Ex Ex Exe Ex

a x̅ bi 67 0.52553 410
39

a bx

aamt coefficient of determination R

8 0 852
R 0.85272 0.726
0.5210 interpretation and R

gH
qq.iq ffE zEgq
increasemonetmysalesbyo.SI
gg qggggy
is explained by number of
a
advertisement
 GUIDING

1 2 3 4 5

se

É
3 12 4
6
3 2
8 5 40 64

9 6
2
301 8 20
1136
21 6

IE
a x̅ bx̅ 4 0.615 6 310
0

a bx
i
Interpretation value of a and b:
b 0.615
to
b 0.615 intecatesThat an

1 83,1 0.615
increase in sales volume by per
increases 0.61s
84 4 1
one unit oft.gs
 I 2 3 4 5 6 7
independt

dependent

e
35 9
E x̅
4
9

71 fi 42

224
f 1521 f
4
225 625
F 210EX 63EX.x
2082EE.FI
b

a x̅ b x̅ 9 0.241 30 1.77C
 2 5

12 55
14 63 882 196

71
I 1 3 61 2
17
16
67
70
1 1139

1120
1289
256
1
11 51 561 121

Iftonsoxenost 10

b 44 61 22
100
1472

a x̅ b x̅ 61 2 314 119.21

Estimate 15years

647W
D 164 481
I ˢ 812
0.056

a x̅ b x̅ 6 0.055 81 6.440

x̅

I
41 486 6
 independent ydependent

005
7196

simple regression

a bx
102.638 0.547239

5 0 547239
inficateThan increase age by one year
an
increaseThe blood pressure
by 0.547239

t fast 1 82 21.96 There is no effect

p value 0.117 70.05 There is no effect

 independent dependent

7196 go05

a
b

simple regression

a bx
71.2 8 Sx

b 8.5 indicateThan an increase effect of

will increaseThe quality by one unit
sales of product by 8.5

fast 6.367 1.96 There is effect

p value 0.0007 0.05 There is effect

 independent independent

tersonsent

4005 71.96www

multiple regression

a
by box
2 76
0.3614 0.196

the
0.54
R 0.54812 0 300

T.SI59aiitxo 3617005
Total cost 1.96 Both 70.05 There is no effect
0.05

Type of effect Sales quality possitive effect

Total cost Negative effect

III
bam
 independent dependent

r
r

7.1.96
c on

dependent y income
independent
investment

Simpleregression

n 9

2 0
8849819
R Co 884 0.781

a
significance f 0.0015
a 0.05 misc airamisig f if in
Not vain a is is 51 Sig f is 11
Significance f 00015
0.05
YES the investment has an
significant effect on Theincome
Piast oz 1.96 There is effect

p value 0.0015 0.05 There is effect

Type ofeffect
positive effect

a bx
I 182 889 0 1089

I
3000
I 182 889 0 1089
182 889 0 1089 3000
x̅ 509
The Slope of the Regression Line and the Correlation Coefficient
The reason for the connection between the value of r and the slope of the least squares
line has to do with the formula that gives us the slope of this line.

For paired data (x, y) we denote the standard deviation of the x data by Sx and the
standard deviation of the y data by Sy.

The formula for the slope an of the regression line is: 𝑏=𝑟

Example:

Given the following data:

 Coefficient of correlation between (X & Y) is (0.9).

 The standard deviation of X is (4), and the standard deviation of Y is (2.5).

 The mean of X is 12 and the mean of Y is 8.

Required:

1) Find the regression equation of Y on X.

2) Estimate the value of ŷ if the value of X is 22.

Solution:
1) r = 0.9, Sx = 4, Sy = 2.5, 𝑥 = 12, 𝑎𝑛𝑑 𝑦 = 8
.
𝑏=𝑟 = 0.9 = 0.5625

𝑎 = 𝑦 − 𝑏𝑥 = 8 - 0.5625 (12) = 1.25

ŷ = a + bx
ŷ = 1.25 + 0.5625x
2) Let X = 22, then ŷ = 1.25 + 0.5625(22) = 13.625

10
b 8 0.9
8 0.9
0.56266
a
x̅ bx̅ 8 0.5626 12
I a bx
I 125 0 5625

22
1.25 0.5625
1.25 0 56266227
13 625
 4 0.8

b r 0.8

A
Y bx̅ 5 0.4 8

i i

17
 R2

2 650 08

b r 0 8
8 ÉI
a
I bx̅ 6 0.571 a I 2 3g
I a bx
0.577
1 0.861 1

20

ii on
12.281
References
 Miah, A.Q. (2016). Applied Statistics for Social and Management Sciences.

Springer, Singapore.

 Cheng-Few Lee, John C. Lee, Alice C. Lee, (2013). Statistics for Business and

Financial Economics, 3rd edition, ISBN: 978-1-4614-5896-8.

12