Upload
norlianah-mohd-shah
View
217
Download
0
Embed Size (px)
Citation preview
7/28/2019 Presentation Chapter 6 MTH1022
1/45
TOPIC 6CORRELATION & REGRESSION
Module Outcomes:
[MO1] Identify basic mathematical concepts, skills and mathematical
techniques for algebra, calculus and data handling.[MO2]Apply the mathematical calculations, formulas, statistical
methods and calculus techniques for problem solving in industry.
[MO3]Analyse calculus and statistical problems in industry.
7/28/2019 Presentation Chapter 6 MTH1022
2/45
7/28/2019 Presentation Chapter 6 MTH1022
3/45
INTRODUCTION
Positive linear relationship:
An increase in one variable cause another variable to
increase
Example: Sugar pr ice and Food pr ice
Negative linear relationship:
An increase in one variable cause another variable to
decrease
Example: Energy price and companies profits
7/28/2019 Presentation Chapter 6 MTH1022
4/45
THESCATTERDIAGRAM
Use to determine whether a relationship exist
between two variables.
The independent variable is normally labeled on the
horizontal axis and dependent variable on the
vertical axis.
7/28/2019 Presentation Chapter 6 MTH1022
5/45
Rectangular coordinate
Two quantitative variables
One variable is called independent (X) and
the second is called dependent (Y)
Points are not joined
No frequency table
Scatter diagram
Y
* *
*
X
7/28/2019 Presentation Chapter 6 MTH1022
6/45
SCATTERDIAGRAMOFWEIGHTANDSYSTOLICBLOOD
PRESSURE
80
100
120
140
160
180
200
220
60 70 80 90 100 110 120
wt (kg)
SBP(mmHg)
Wt.
(kg)
67 69 85 83 74 81 97 92 114 85
SBP
mHg)
120 125 140 160 130 180 150 140 200 130
7/28/2019 Presentation Chapter 6 MTH1022
7/45
80
100
120
140
160
180
200
220
60 70 80 90 100 110 120Wt (kg)
SBP(mmHg)
Scatter diagram of weight and systolic blood pressure
7/28/2019 Presentation Chapter 6 MTH1022
8/45
No relationship between weight and pulse rate
7/28/2019 Presentation Chapter 6 MTH1022
9/45
THE SCATTER PLOTS
The pattern of data is indicative of the type of
relationship between your two variables:
positive relationship
If x increase, y will increase as well. negative relationship
If x increase, y will decrease and if x decrease, y will
increase.
no relationship
If the slope is zero, so the two variable are not
related.
7/28/2019 Presentation Chapter 6 MTH1022
10/45
POSITIVERELATIONSHIP
7/28/2019 Presentation Chapter 6 MTH1022
11/45
0
2
4
6
8
10
12
14
16
18
0 10 20 30 40 50 60 70 80 90
Age in Weeks
HeightinCM
7/28/2019 Presentation Chapter 6 MTH1022
12/45
NEGATIVERELATIONSHIP
Reliability
Age of Car
7/28/2019 Presentation Chapter 6 MTH1022
13/45
NORELATION
7/28/2019 Presentation Chapter 6 MTH1022
14/45
7/28/2019 Presentation Chapter 6 MTH1022
15/45
Ex: Draw a scatter diagram for the following data and state the type of
relationship between the variables.
P 3 4 6 8 11 17 18 22q 10 20 3 7 4 1 12 5
0
5
10
15
20
25
0 5 10 15 20 25
q
q
Based on the scatter diagram, it is determined that there is no relationship between
the two variables.
7/28/2019 Presentation Chapter 6 MTH1022
16/45
WHATISCORRELATION???
Correlation analysis is a statistical method used to
measure the STRENGTH of the relationship
between two variables which are independent
variable (X) and dependent variable (Y).
7/28/2019 Presentation Chapter 6 MTH1022
17/45
7/28/2019 Presentation Chapter 6 MTH1022
18/45
Pearsons correlation coefficient is used to measure
the strength of the relationship between two variablesthat are quantitative in nature
This method is not suitable for qualitative data,
especially rank data
7/28/2019 Presentation Chapter 6 MTH1022
19/45
SIMPLE CORRELATIONCOEFFICIENT () It is also called Pearson's correlation or product
moment correlation coefficient.
It measures the nature and strength between two
variables of the quantitative type.
7/28/2019 Presentation Chapter 6 MTH1022
20/45
The sign ofrdenotes the nature of
association
while the value ofrdenotes the
strength of association.
7/28/2019 Presentation Chapter 6 MTH1022
21/45
If the sign is +ve this means the relation isdirect (an increase in one variable is associated
with an increase in the other variable and adecrease in one variable is associated with adecrease in the other variable).
While if the sign is -ve this means an inverse orindirect relationship (which means an increasein one variable is associated with a decrease inthe other).
7/28/2019 Presentation Chapter 6 MTH1022
22/45
The value of r ranges between ( -1) and ( +1)
The value of r denotes the strength of theassociation as illustrated
by the following diagram.
-110
-0.25-0.75 0.750.25
strong strongintermediate intermediateweak weak
no relation
perfect
correlation
perfect
correlation
Directindirect
7/28/2019 Presentation Chapter 6 MTH1022
23/45
Ifr= Zero this means no association orcorrelation between the two variables.
If0 < r< 0.25 = weak correlation.
If0.25 r< 0.75 = intermediate correlation.
If0.75 r< 1 = strong correlation.
Ifr= l = perfect correlation.
7/28/2019 Presentation Chapter 6 MTH1022
24/45
Example : The table below show the interest rates for car loans and the average
number of customers who apply for the loans in a month from a finance company.
Interest rate
in %, x 6.0 6.2 6.5 6.8 7.0
Number of
applicants, y80 80 78 75 70
x y xy6.0 80 36 6400 480
6.2 80 38.44 6400 496
6.5 78 42.25 6084 507
6.8 75 46.24 5625 510
7.0 70 49 4900 490
= 3 2 . 5 = 3 8 3 =211.93 =29409 =2483
7/28/2019 Presentation Chapter 6 MTH1022
25/45
EXAMPLE:
Weight
(Kg)
Age
(years)
serial
No
1271
862
1283
1054
1165
1396
A sample of 6 children was selected, data about their
age in years and weight in kilograms was recorded as
shown in the following table . It is required to find the
correlation between age and weight.
7/28/2019 Presentation Chapter 6 MTH1022
26/45
These 2 variables are of the quantitative type, one variable
(Age) is called the independent and denoted as (X)
variable and the other (weight) is called the dependent and
denoted as (Y) variables to find the relation between ageand weight compute the simple correlation coefficient
using the following formula:
7/28/2019 Presentation Chapter 6 MTH1022
27/45
Y2X2xy
Weight
(Kg)
(y)
Age
(years)
(x)
Serial
n.
14449841271
643648862
14464961283
10025501054
12136661165
169811171396
y2=
742
x2=
291
xy=
461
y=
66
x=
41
Total
7/28/2019 Presentation Chapter 6 MTH1022
28/45
r = 0.759
strong direct correlation
7/28/2019 Presentation Chapter 6 MTH1022
29/45
EXAMPLE: RELATIONSHIPBETWEEN ANXIETYAND
TEST SCORES
Anxiety
(X)
Test score
(Y)X2 Y2 XY
10 2 100 4 20
8 3 64 9 242 9 4 81 18
1 7 1 49 7
5 6 25 36 306 5 36 25 30
X = 32 Y = 32 X2 = 230 Y2 = 204 XY=129
7/28/2019 Presentation Chapter 6 MTH1022
30/45
CALCULATING CORRELATION COEFFICIENT
94.0
)200)(356(
1024774
32)204(632)230(6
)32)(32()129)(6(22
r
r = - 0.94
Indirect strong correlation
7/28/2019 Presentation Chapter 6 MTH1022
31/45
=0.09
Pearsons product moment correlation coefficient is -0.09. since this value isclose to -1.0, we conclude that there is a very strong negative linear
relationship between variable x and y.
7/28/2019 Presentation Chapter 6 MTH1022
32/45
7/28/2019 Presentation Chapter 6 MTH1022
33/45
=0.953
Pearsons product moment correlation coefficient is 0.953. since this value
is close to 1.0, we conclude that there is a very strong positive linear
relationship between variable x and y.
7/28/2019 Presentation Chapter 6 MTH1022
34/45
SPEARMANS RANK CORRELATION COEFFICIENT
1. Spearmans rank correlation coefficient is a measure ofassociation between two variables that are at least of
ordinal scale.
2. Suitable for qualitative data and quantitative data but
quantitative data must transformed into qualitative data.
= 1 6
( 1)To compute Spearmans rank correlation:
1. Make a list of the n subject ( or observation)
2. Enter the rank of X and rank of Y variable.
3. Find = .4. Find .
7/28/2019 Presentation Chapter 6 MTH1022
35/45
Example:
Six models of cars were rated for style and performance by customers at two
different locations. The table below shows the average ratings by the customers.
The scale is from 1 to 100.
Car model J K L M N P
Location 1 50 78 32 90 63 42
Location 2 40 49 33 55 80 33
Is there a relationship between the two ratings. Prove by calculating the
Spearmans rank correlation coefficient.
7/28/2019 Presentation Chapter 6 MTH1022
36/45
Car
model
Location 1
x
Location 2
y
Rank of x
Rank of y
= - ()
J 50 40 3 3 0 0
K 78 49 5 4 1 1
L 32 33 1 1.5 -0.5 0.25
M 90 55 6 5 1 1
N 63 80 4 6 -2 4
P 42 33 2 1.5 0.5 0.25
=6.5
=1- (.)()
= 1- = 0.81 = 1
6 ( 1)
Spearmans rank correlation coefficient is 0. since this value is close to 1.0, we
conclude that there is a very strong positive linear relationship between variable
x and y.
7/28/2019 Presentation Chapter 6 MTH1022
37/45
Example :
In a singing competition, two judges ranked the seven finalists as shown in the
table below. Compute spearmans rank correlation coefficient of the two sets of
rating. What can be said about the ratings of the two judges?
Finalists A B C D E F
Judge 1 5 6 3 9 4 8
Judge 2 3 4 1 8 5 10
7/28/2019 Presentation Chapter 6 MTH1022
38/45
Finalists Judge 1
x
Judge 2
y
Rank of x
Rank of y
()
A 5 3 3 2 1 1
B 6 4 4 3 1 1C 3 1 1 1 0 0
D 9 8 6 5 1 1
E 4 5 2 4 -2 4
F 8 10 5 6 -1 1
G 10 11 7 7 0 0
=8
=1- ()()
= 1-
= 0.86 = 1 6
( 1)Spearmans rank correlation coefficient is 0.86. since this value is close to 1.0,
we conclude that there is a very strong positive linear relationship between
variable x and y.
7/28/2019 Presentation Chapter 6 MTH1022
39/45
= +
The method to find the equation is by using Least Squares Method.
General equation:
= ( ) =
Where:
n is the number of data
7/28/2019 Presentation Chapter 6 MTH1022
40/45
7/28/2019 Presentation Chapter 6 MTH1022
41/45
Example:
Find the least squares regression line of y on x for the following data.
X 3 6 9 11 16 18
y 2 8 11 14 19 21
Determine the values of y when (a) x = 5 and (b) x = 14
7/28/2019 Presentation Chapter 6 MTH1022
42/45
x y xy3 2 9 4 6
6 8 36 64 48
9 11 81 121 9911 14 121 196 154
16 19 256 361 304
18 21 324 441 378
= 6 3 = 7 5 =827 =1187 = 9 8 9
The regression line is y = a + bx and
=
( )
= 6 989 (63)(75)6 827 (63) =
= 1.218
=
= 756 1.218 636 = - 0.289
Thus, the regression line is y = - 0.289+1.218x
When x = 5, y = - 0.289+(1.218 5 ) = 5.801When x = 14, y = - 0.289+(1.218 14 ) = 16.763
7/28/2019 Presentation Chapter 6 MTH1022
43/45
Example : The table below gives the pairs of measurements for two variables x
and y.
X 5 9 12 15 18 20
y 3 5 8 9 10 12
Find the regression line of y on x using the least squares method. Estimates thevalue of y when x = 11.
7/28/2019 Presentation Chapter 6 MTH1022
44/45
x y xy 5 3 15 25 9
9 5 45 81 25
12 8 96 144 6415 9 135 225 81
18 10 180 324 100
20 12 240 400 144
= 79 = 47 = 711 = 1199 = 423
= ( )
= ()()
()
=
= 0.58
=
= 47
60.58 79
6
= 7.63 7.637
= 0.196
Thus, the regression line is y = 0.196+0.58x
When x = 11, y = 0.196+ (0.58 11)=6.679
7/28/2019 Presentation Chapter 6 MTH1022
45/45
THE ENDPlease Do a Lot of ExercisesTHANK YOU
https://mth1022icam.blogspot.com
JUST REMEMBER: