UNIT 3: SUMMATION NOTATION, MEASURES OF CENTRAL TENDENCY AND DISPERSION

3.0 AIMS AND OBJECTIVE

At the end of this unit, you will be able to:
Apply properties of summation notation to solve problems which involve the summation
operator.Understand the definition and properties of central tendency and dispersionCompute problems
of the central tendency and dispersion for ungrouped and grouped data.Identify the advantages and
application areas of the central tendency and dispersion.
3.1 SUMMATION NOTATION AND ITS PROPERTIES
operator, , implies that the values that follow it are to be summed or added together.
Summation operator,
n → upper limit
∑ x i → the ith variable of x
i = m→ lower lim it
5
∑ xi
Example i=1 = x1 + x2 + x3 + x4 + x5
Properties:
1. The summation of sums of differences
n n n n n n
∑ ( x i + y i ) = ∑ xi + ∑ yi ∑ ( x i − y i ) = ∑ xi − ∑ y i
i=1 i=1 i=1 , i=1 i=1 i=1

Example: Suppose x1 = 1 , x2 = 3 , x3 = 4 , y1 = 2 , y2 = 5 , y3 = 3
3 3 3
∑ ( xi + yi) = ∑ xi + ∑ yi
Then i=1 i=1 i=1

(x1+ y1) + ( x2 + y2) + ( x3 + y3) = (x1 + x2 + x3) + ( y1 + y2 + y3)

(1 + 2) + ( 3 + 5) + ( 4 + 3) = (1 + 3 + 4) + ( 2 + 5 + 3)

3 + 8 + 7 = 8 + 10

18 = 18
3 3 3
∑ ( xi − y i ) = ∑ xi − ∑ yi
i=1 i=1 i=1 …… left for the student

2. Multiplication by a constant
n n
∑ kx i = k ∑ x i
i=1 i=1
Example: Suppose k = 7 and x1 = 2 , x2 = 5 , x3 = 4
3 n
∑ kx i = k ∑ x i
Then i=1 i=1

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 1
kx1 + kx2 + kx3 = k(x1 + x2 + x3)

7(2) + 7(5) + 7(4) = 7(2 + 5 + 4)

14 + 35 + 28 = 7(11)
77 = 77
3. Summation of a constant
n
∑ k = nk
Case 1:
1: If lower limit equal to 1, i=1
Example: suppose k = 6 and upper limit = 4
4
∑ k = nk
Then i=1
4
∑ 6 = 4 ( 6)
i=1
5 + 6 + 6 + 6 = 24
4. Sum of summations
k n n
∑ xi + ∑ xi = ∑ xi
i=1 i=k +1 i=1 for any k < n

Example:
Example: Suppose x1 = 2 , x2 = 4 , x3 = 6, x4 = 3, x5 = 2, x6 = 4 let k = 3 then
k n n
∑ xi + ∑ xi = ∑ xi
i=1 i=k +1 i=1

3 6 6
∑ xi + ∑ xi = ∑ xi
i=1 i=4 i=1

(x1 + x2 + x3) + (x4 + x5 + x6) = x1 + x2 + x3 + x4 + x5 + x6

(2 + 4 + 6) + (3 + 2 + 4) = (2 + 4 + 6 + 3 + 2 + 4)
21 = 21
6 6
∑ x i = 10 , ∑ x i2 = 148
Leti=3 i=3 , x1 = 3 , x2 = 2
6 6 6
∑ xi ∑ x i2 ∑ xi( xi − 2 )
Find i. i=1 ii. i =1 iii. i=1

6 2
∑ (2 x i − 3 ) 2
∑ ( ix i + 4 )
i=1 v. i=1

iv.

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 2
3.2 MEASURES OF CENTRAL TENDENCY

3.2.1 ARITHMETIC MEAN (AM)

Definition 3.2.1

The arithmetic mean is the sum of the values in a group divided by the number of items in that group.
Let x1, x2, …,xn be n values of a variable x, then their arithmetic mean is defined by:
n
∑ xi
x̄ =
x1 + x2 + ⋯ + xn
= i =1
=
∑x
n n n
Where x – sum of all observations, n – total number of observations

Computation Of Arithmetic Mean For Ungrouped And Grouped Data

For ungrouped data:

n
∑ xi ∑d
i=1
x̄ = x̄ = A +
Direct method: n Short cut method: n
Where n – number of items A = Assumed mean d = sum of deviations i.e. ( xi - A)

Example:
Example: Find the arithmetic mean for the following data by
i. direct method ii. short cut method

23.4 15.6 22.1 20.0 26.7 31.4 18.9 22.3

Solution:
8
8 ∑ xi
∑ xi i=1 180 . 4
x̄ = = = 22. 55
i. i=1 = 180.4 , n = 8 n 8

ii. Let A = 22 then di : 1.4, -6.4, 0.1, -2, 4.7, 9.4, -3.1, 0.3
8
8 ∑ di
∑ di i=1 4.4
x̄ = A + = 22 +
i=1 = 4.4 , n = 8 n 8 = 22 + 0.55 = 22.55
For grouped data:For
data:For Discrete
n
∑ f i xi ∑ fd
x̄ = i=1
=
∑ fx x̄ = A ±
Series: Direct method: n n Short cut method: n
Where f - frequency d - deviation of items from assumed mean (xi – A)
A - assumed mean n - number of observations

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 3
 Example:Given
Example:Given data of 50 students of marks of a test in a class. Calculate the arithmetic mean by
i. direct method ii. short cut method.

No. of Students 20 30 40 50 60 70
Marks 8 10 16 8 5 3

Solution: Marks xi fi fx Di = ( x – 40) fd

where A = 40
i 20 8 160 -20 -160
30 10 300 -10 -100
40 16 640 0 0
50 8 400 10 80
60 5 300 20 100
70 3 210 30 90
50 2010 10

x̄ =
∑ fx 2010
= = 40 . 20
n 50
x̄ = A ±
∑ fd = 40 + 10 = 40 . 20
ii. n 50
For continuous series:

Direct method
x̄ =
n
∑ fcm i
Step deviation method
x̄ = A ±
n ( )
∑ fd1 ×c

Where , f – frequency n – number of observation

Cmi – class mark A – assumed mean
d – derivation of class marks from assumed mean (cmi – A)
d' – d/c c – class width
Example: In a survey, the number of persons at different ages is found as follows:

Age in Year 5 - 15 15 - 25 25 - 35 35 - 45 45 - 55 55 - 65
No. of Persons 8 10 14 20 16 12

Classes f cmi fcm d = cm – A where A df d1 = d/c fd1

= 30
5 - 15 8 10 80 -20 -160 -2 -16
15 - 25 10 20 200 -10 -100 -1 -10
25 - 35 14 30 420 0 0 0 0
35 - 45 20 40 800 10 200 1 20
45 - 55 16 50 800 20 320 2 32
55 - 65 12 60 720 30 360 3 36
80 3020 620 62

Solution:

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 4
 x̄ =
∑ fcm 3020
= = 37 . 75
i. n 80

x̄ = A +
∑ fd = 30 + 620 = 30 + 7 .75 = 37 .75
ii. n 80

x̄ = A + ( )
∑ fd 1
n
× c = 30 + ( )
62
80
10 = 30 + ( 0. 775 ) 10 = 37 .75

iii.
3.2.1.1 Properties of the Arithmetic Mean

The mathematical properties of an arithmetic mean are discussed below:

1) The product of arithmetic mean ( x̄ ) and the number of observations (N) on which the mean
n
x̄ N = ∑ xi
based is equal to the sum of all given values i.e. i=1
2) The algebraic sum of the deviations of the given values from the arithmetic mean is equal to
zero. Mathematically,
∑ ( x i − x̄ ) = 0 … for ungrouped data
∑ f i ( x i − x̄ ) = 0 … For grouped data
Because of this property, the arithmetic mean may be characterized as a point of ‘Balance’

3) Suppose the mean of the values x1 , x2, … , xn be

x̄ 0 . Then
i.
x̄
if a constant k is added to each xi, then the new mean n will be 0 + k.
x̄
ii. if a constant k is subtracted from each xi then the new mean of xn will be x0- k
iii. if if a constant k is multiplied by each xi then the new mean of xn will be kx0
iv. if each xi is divided by constant k then the new mean of xn will be x0/k
Example :if xi is 2,4,6,8,10 and k=2then
k=2then find the mean of
a) xi + k b) xi - k c)kxid) xi/k
4
Example: Given data 12, 10, 8, 6, 16, 7, 11. If each item is multiplied by 5 and 8 is added, what will
be the new mean?

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 5
 4
x̄ n = x̄ + 8
5 0
7
∑ xi 4
i=1 70 x̄ n = ( 10 ) + 8 = 16
x̄ 0 = = = 10
7 7 New mean 5

Ex. Given data 3, 8, 9, 4, 7, 5, 10, 11, 6 if each item is multiplied by 2 and 6 is added, then
i. The new mean will be _______________
ii. ∑ ( x i − x̄ ) = __________________

Combined (Pooled) Arithmetic Mean

Let x̄ 1 and x̄ 2 be the AM of two groups, having N1 and N2 observations respectively. If we combine the
two groups in to a single group, then the arithmetic mean of the combined group will be
n

N 1 x̄ 1 + N 2 x̄ 2 + … + N n x n ∑ N i x̄ i
i=1
x̄ c = =
N 1 x̄ 1 + N 2 x̄ 2 N 1 + N2 + … + N n n
x̄ c =
N 1 + N2 ∑ Ni
For n number of groups, i =1

Example:The
Example:The mean height of 25 male and 20 female is 161.0cm and 155.6cm. What will be the
combined mean height?
x̄ m = 161.0cm, x̄ F = 155.6cm, N = 25, N = 20
M F
x̄ m N m + x̄ F N F
x̄ c =
Nm + NF
161 . 0 ( 25 ) + 155. 6 ( 20 ) 7137
x̄ c = = = 158 . 60 cm
25 + 20 45

Ex. 4 In a factory, 120 workers get an average wage of birr 30 a day, 160 workers get Birr 50 a day, 80
workers get Birr 60 a day and 40 workers get birr 80 a day. Find
i. the average of averages.
ii. the general average.

3.2.1.2 Weighted Arithmetic Mean

An item or value may be relatively more important or less important than other items. This relative
importance is technically known as weight. In case where the relative importance of the different items
is not the same we compute weighted arithmetic mean.

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 6
 If w1, w2, …, wn are weights attached to the values x 1, x2, … , xn respectively, then the weighted AM is

x̄ w =
x 1 w1 + x 2 w2 + … + x n wn
=
∑ wx
w 1 + w2 + … + w n ∑w
defined as
Example: An auto ride costs Birr 5 for the first km, Birr 4 for the next 3kms and Birr 9 for each of the
subsequent kms. Find the average cost per km for 10 kms.

Rate (Birr) Distance (km) w xw

5.00 1 5.00
4.00 3 12.00
9.00 6 54.00
10 71.00

x̄ w =
∑ xw =
71 .00
= 7 . 10 Birr
∑w 10

EX. Given data

Designation Class I Class II Subordinate Clerical Lower

Officer Officer Staff Staff Staff
Monthly Salary (in birr) x 1500 800 500 250 100
Strength of Cadre. w 10 20 70 100 150

Calculate i. the simple arithmetic mean.

ii. the weighted arithmetic mean.

Correcting the Arithmetic Mean

Wrong Sum − Wrong Entry + Correct Entry

Correct Mean =
Formula:
Formula: Total Number of Observations

Examples:

1. The average mark of 100 students was found to be 40 but latter it is discovered that a score of 33
was misread as 83. Find the correct average corresponding to the correct sum.

x̄ = 40¿ }¿¿→∑ xi = x̄ N = 40 ( 100 ) = 4000¿ wrong sum

Wrong Entry = 83
Correct Entry = 33
4000 − 83 + 33 3950
Correct Mean = = = 39 .5
100 100

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 7
2. The average of a class having 35 pupils is 14 years. When the age of the class teacher is added to
the sum of the ages of the pupils, the average rises by 0.5 year. What must be the age of the teacher?

x̄ = 14¿ }¿ ¿→∑ xi = 14 ( 35) = 490 ¿ … Sum of ages of the pupils

x̄ = 14.5¿} ¿¿→∑ xi = 14.5 ( 36 ) = 522¿ … Sum of ages of the pupils and the teacher
 Age of the teacher is 522 - 490 = 32 years.
3. Goals scored by a football team in successive matches are 5, 2, 4, 3, 6, 0, 4 and 6. What is the
number of goals the team must score in the next match in order that the average comes to 4 goals per
match?

Total goal scored in 8 matches = 5 + 2 + 4 + 3 + 6 + 0 + 4 + 6 = 30

Total goal scored in 9 matches = x̄ .N = 4  9 = 36
Hence the goals required in the 9th match to bring the average 4 = 36 – 30 = 6

3.2.2 MODE

A value, which occurs most frequently in a series of observations, is called Mode. So by looking the
observations mode can be identified.
A data may or may not have mode and may have more than one mode .

Computation of Mode for Ungrouped and Grouped Data

data: Mode ( x^ ) = that value in the data set, which occurs most often.
For ungrouped data:

data: Discrete Series: Mode ( x^ ) = the value of the variable corresponding to the
For grouped data:
maximum frequency.

Continuous Series:
Series: The class corresponding to the maximum frequency is called the modal class. The
value of mode is obtained by the following interpolation formula

Mode ( ^x ) = l + 1
f − f0
[
( f 1 − f 0) + ( f 1 − f 2 )
× c
]
or

( )
Δ1
Mode ^x = l + c
Δ1 + Δ 2
Where l – LCB of the modal class f2 – frequency succeeding f1
f1 – maximum frequency C – magnitude of the class
f0 – frequency preceding f1 ∆1 = f1 – f0
f2 – frequency succeeding f1 ∆2 = f1 – f2

Example:Find
Example:Find the value of mode for the following data

i. 25, 15, 23, 40, 27, 25, 23, 25, 20, 19, 22, 24, 25
ii.

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 8
 x 10 20 30 40 50 60
f 4 9 16 25 22 15

iii.
Classes 0-9 10 - 19 20 - 29 30 - 39 40 - 49 50 - 59 60 - 69 70 - 79
fi 328 350 720 664 598 524 378 244

Solution:

i. Mode = value which occurs most often

Mode = 25

ii. Mode = Value of the variable with maximum frequency

Mode = 40

iii. Modal Class = 19.5 - 29.5

l = 19.5 f0 = 350 f1 = 720 f2 = 664 c = 10

Mode ( ^x ) = l +
[ f1 − f0
( f 1 − f 0) + ( f 1 − f 2 ) ]
× c

= 19 .5+
[ 720 − 350
( 720 − 350 ) + ( 720 − 664 ) ] × 10 = 19 . 5 +
3700
426
= 28 .1854

EX Find the modal value of the following data.

i. 27, 33, 42, 25, 23, 27, 25, 33, 27, 28, 16, 18

ii. Height (in inches) 58 60 61 62 63 65 68 70

No. of Persons 4 6 10 8 20 24 9 5

iii.
Classes 0 - 400 400 - 800 800 - 1200 1200 - 1600 1600 - 2000
fi 4 12 40 41 27

3.2.3 MEDIAN

The median is that value of the variable, which divides the group in to two equal parts, one part
comprising all the values greater and the other all the values less than median. Or median can be

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 9
 defined as the middle value of a set of data values when they are arranged in ascending or
descending order.
Commutation of Median for Ungrouped and Grouped Data
For ungrouped data:
data:
First, rearrange the values in the order of magnitude.
Then apply the following formula.

( )
th
N +1
Median (~
x ) = vallue of the item ( where n is odd)
2
xn + 1
=
2

[ ( ) ( ) ]
th th
1 N N
Median (~
x) = Value of item + Value of + 1 item ( Where n is even )
2 2 2
1
=
[
x + xn
2 n2 2
+1 ]
For grouped data:
Discrete Series:
Series:
1. Compute the <cfi column.
N +1
2. Search for the value of 2 , if not available, consider the value just greater than it, in the
column of <cfi.

Continuous Series:
Series:

1. Compute the <cfi column.

N
2. Search for the value of 2 if not available, consider the value just greater than it, in the column
of <cfi.
3. Then the following interpolation formula is used to calculate the median.
Median (~
x) = l +
c
f ( n2 − c . f )
where l- LCB of the median class
c - class interval of the median class
f - frequency of the median class
N
c.f- cumulative frequency jut less than 2
Example: Find the median of the following data.

i. a) 27, 33, 42, 25, 23, 25, 33, 28, 27, 16, 18, 12

b) 8, 5, 2, 6, 15, 10, 25
x 4 6 8 10 12 14 16
Statistics forf Finance2lecture note
4 by Legese
5 T. July
3 2009/2016
2 edtied4 1 Page 10
ii.

iii.
x 50 - 60 60 - 70 70 - 80 80 - 90 90 - 100 100 - 110
fi 20 21 50 40 53 16

Solution:

i. a. Rearranging:
12 16 18 23 25 25 27 27 28 33 33 42
n = 12 … even

~ 1 1 1
x = x + xn
2 n2 2 [ + 1 ] =
2
[ x 6 + x 7] =
2
[ 25 + 27 ] = 26

b. Rearranging: 2 5 6 8 10 15 25
n = 7 … odd
~ x
x = n+1 = x 4 = 8
2
ii.
x 4 6 8 10 12 14 16
f 2 4 5 3 2 4 1
<cfi 2 6 11 14 16 20 21

n = 21 Median = The value of N + 1 th item

2
th
= 21 + 1 item
2
= The value of the 11th item
= 8
iii.
x 50 - 60 60 - 70 70 - 80 80 - 90 90 - 100 100 - 110
fi 20 21 50 40 53 16
<cfi 20 41 91 131 184 200

()
th
n
item = 100th item = 80 - 90
Median class = Value of 2
l = 80, c = 10, f = 40, c.f = 91
Median (~
x) = l +
c n
f 2
− c . f = 80 +
10
40( )
( 100 − 91 ) = 80 +
9
4
= 82 . 25
EX Find the median of the following data.

i. a. 20 15 21 13 22 24 22 25 26 27 25

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 11
 b. 120 125 112 137 129 127

ii.
x 28 30 32 34 36 38 40 42
f 14 15 16 24 16 10 6 4

iii.
x 30 - 34 35 - 39 40 - 44 45 - 49 50 - 54 55 - 59
fi 5 10 15 20 6 4

3.2.4 QUARTILES, DECILES AND PERCENTILES

Definitions:

Quartiles: are the three values, which divide the given data in to four equal parts. They are denoted by
Q1, Q2 and Q3.
Q1 - The lower or first quartile. It covers 25% of the distribution.
Q2 - The middle or second quartile. It covers 50% of the distribution.
Q3 - The upper or third quartile. It covers 75% of the distribution.

Deciles: are the nine values, which divide the series in to ten equal parts. They are denoted by D1, D2,
… , D9.
D1 = Covers 10% of the distribution
D2 = Covers 20% of the distribution
.
.
.
D9 = Covers 90% of the distribution
Percentiles:
Percentiles: are the 99 values, which divide the series in to 100 equal parts. They are denoted by P1, P2
, … , P99.
Note that: i. Q1 = P25 Q2 = D5 = P50 = median Q3 = P75
ii. D1 = P10, D2 = P20, D3 = P30, … , D9 = P90.

Importance:

The quartiles are more widely used in Economics and Business while the deciles and percentiles are important in
Psychology and Educational Statistics concerning grades, rates, ranks, etc. The working principle for computing
the partition value is basically the same as that of computing the median.

Computation of Quartiles, Deciles and Percentiles for Ungrouped and Grouped Data

For ungrouped data and discrete series:

First, for ungrouped data, rearrange the values in the order of magnitude and for discrete series, compute the
<Cficolumn. Then apply the following formula.

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 12
 i ( N + 1 )th
Qi = value of item
4
i ( N + 1 )th
Di = value of item
10
i ( N + 1 )th
Pi = value of item
100
For continuous series:
1. Compute the <cficolumn.
2. Determine the quartile, decile or percentile class.
3. Apply the following interpolation formula.

Qi = l +
c
f ( iN4 − c . f )
Di = l +
c
f ( iN10 − c . f )
Pi = l +
c
f ( iN100 − c . f )
Example: For the data given below, compute the value of Quartiles, D3, D5D7, P50 and P75 and interpret.

a) 1,2,3,4,5,6,7,8,9,10,and 11
b)
Marks Below 10 10 - 20 20 - 40 40 - 60 60 - 80 Above 80
No. of Students 10 15 25 30 14 6
<cfi 10 25 50 80 94 100

Solution:b
N th
Q1 – size of 4 item = 25th item 10 – 20 quartile class
l = 10, c = 10, f = 15, c.f = 10

Q1 = l +
c
f ( n4 − c . f ) = 10 + 1015 ( 25 − 10 ) = 20
Mark of 25% of students is less than 20.
2N th
Q2– size of 4 item = 50th item 20 – 40 quartile class
l = 20, c = 20, f = 25, c.f = 25

Q2 = l +
c
f ( n2 − c . f ) = 20 + 2025 ( 50 − 25 ) = 40
Mark of half of students is below 40.
3N th
Q3– size of 4 item = 75th item 40 – 60 quartile class
l = 40, c = 20, f = 30, c.f = 50

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 13
 Q3 = l +
c
f ( 34n − c . f ) = 40 + 2030 ( 75 − 25 ) = 73 .33

3
Mark of 4 thof students is below 73.33.
3N th
D3– size of 10 item = 30th item 20 – 40 decile class
L = 20, c = 20, f = 25, c.f = 25

D3 = l +
c
f (103 n − c . f ) = 20 + 2025 ( 30 − 25 ) = 24
Mark of 30% of students is below 24.
7N th
D7– size of 10 item = 70th item 40 – 60 decile class
L = 40, c = 20, f = 30, c.f = 50

D7 = l +
c
f ( 7n
10 )
− c . f = 40 +
20
30
(70 − 50 ) = 53 .33
Mark of 70% of students is below 53.33.
15N th
P15– size of 100 item = 15th item 10 – 20 percentile class
L = 10, c = 10, f = 15, c.f = 10

P15 = l +
c 15 n
f 100 (
− c . f = 10 +
10
15 )
( 15 − 10 ) = 13 . 3
Mark of 15% of students is below 13.3.
88N th
P88– size of 10 item = 88th item 60 – 80 percentile class
L = 60, c = 20, f = 14, c.f = 80

P88 = l +
c 88 n
f 100 (
− c . f = 60 +
20
14 )
( 88 − 80 ) = 71. 43
Mark of 88% of students is below 71.43.
EX Compute the value of Quartiles, D4,P69 and interpret for the data given below.

i. 46 35 28 52 54 43 35 49 46 50 41

ii. Daily Wages 40 45 50 55 60 65 70

No. of Workers 9 22 26 18 13 8 5
iii
Rent in 150-250 250-350 350-450 450-550 550-650 650-750 750-850 850-950
Birr
No. of 8 10 15 25 40 20 15 7
Houses

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 14
3.3: MEASURES AND TYPES OF DISPERSION

3.3.1 RANGE

Range (R) can be defined as the absolute difference between the maximum value (Xmax) and the
minimum value (X min) in a given distribution.

Symbolically
R= Xmax - Xmin or R = L - S
Where, R = Range
L = Xmax = maximum value or largest value
S = Xmin= minimum or smallest value.

Range is an about measure of dispersion. Such measures are compared if the unit of measurement is
homogeneous.(i.e if all the sets of distribution are expressed in the same statistical unit such as birr,
letter, meters, etc. But if different sets of distributions are expressed in different statistical units, the
absolute measure of range can not be effectively compared .For .For ensuring comparability, relative
measure of range, known as coefficient of range is used. This is obtained by dividing the absolute range
by the sum of the largest and smallest values in the distribution. Its calculation follows the formula.
x max − x min L −S
=
x
Coefficient of Range max + x min L + S

Discrete Series
Example2: Calculate range and its coefficient from the following data relating to Salaries of 400
Lectures in Unity University.

Salary 1800 2000 2400 2800 3000 3200 3400 4000 4500 5000 5400 7000
(In birr)
Number of 20 25 26 27 30 30 20 40 24 24 94 40
Lectures

Solution:
Range =Xmax - Xmin
=7000-1800
=5200
=5200 Birr
x max − x min 7000 −1800 5200
= = birr = 0 . 591
Coefficient of Range = x max + x min 7000 + 1800 8800

Coefficient of range is high in the distribution. Hence it is more variable and less consistent.
Continuous series

Example 3: Calculate range and its coefficient fro the mark obtained by distance students in statistics.
Mark 11-20 21-30 31-40 41-50 51-60 61-70
Number of Students 15 28 37 10 48 63

Solution: maximum mark is 70 and minimum mark is 11.

Xmax = 70

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 15
 Xmin =11
Range = Xmax- Xmin
= 70 - 11
= 59
x max − x min 70 −11 59
= = = 0. 728
Coefficient of Rage = x max + x min 70 + 11 81

The coefficient of range of marks in statistics in 0.723, hence the marks are more variable. Sometimes,
range may also be calculated from the mid-value of the classes. In such cases, first, mid values are
obtained and then range is calculated by subtracting smallest the mid-value from the largest mid-value.

Example4 From the following distribution, compute range and its coefficient by taking themid-value

Class 5-12 13-20 21-28 29-36 37-44

Frequency 10 9 6 8 4

Solution: Calculation of range and coefficient of range.

Class Frequency Mid-value

5 - 12 10 5 + 12 17
= = 8.5
2 2
13 - 20 9 13 + 20 33
= = 16 . 5
2 2
21 - 28 6 21 + 28 49
= = 24 .5
2 2
29 - 36 8 29 + 36 65
= = 32. 5
2 2
37 - 44 4 37 + 44 81
= = 40 .5
2 2
Range X max - Xmin
Xmax = 40.5
Xmin = 8.5

Range =Xmax- Xmin

=40.5 - 8.5
=32
=32
x max − x min 40 .5 − 8. 5 32
= = = 0 . 653
Coefficient of Range
x max + x min 40 .5 + 8. 5 49
The coefficient of range is large, Hence, the distribution is more variable .

3.2.4 MEAN DEVIATION

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 16
Mean deviation is obtained by dividing the sum of the absolute deviations taken from the average by the
total number of observations. Generally, the result is in the absolute value to denote that deviations are
taken by ignoring algebraic signs. The formula of mean deviation, when deviations are measured from
arithmetic mean is:-

MD x =
∑|d x| = ∑|x i − x|
n n When the data is ungrouped

Where, MD x = mean deviation form mean

Examples 1 Calculate mean deviation from mean for the following data.

X 2 4 6 8 10
F 1 4 6 4 1
Solution Calculation of mean deviation

x  x Deviation from F/D x̄ /

mean /D x̄ /
2 1 2 4 4
4 4 16 2 8
6 6 36 0 0
8 4 32 2 8
10 1 10 4 4
 = 16 x
x = 96 /D x̄ / = 24
/

x̄ =
∑ fx =
96
= 6
N 16

M . D x̄ =
∑ f |D x̄| = 24 = 1 .5
N 16
Example-2 calculates mean deviation and its coefficient from the following continuous series given on
age distribution of 100 high school teachers.

16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 56-60

9 16 12 10 16 14 12 6 5

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 17
Solution of mean deviation and coefficient of mean deviation

 ~
Age group Mid point .x Mean deviation /D x /
x from the mean /D
~
x/
16-20 9 18 162 17.9 161.1
21-25 16 23 368 12.9 206.4
26-30 12 28 336 7.9 94.8
31-35 10 33 330 2.9 29.0
36-40 16 38 608 2.1 33.6
41-45 14 43 602 7.1 99.4
46-50 12 48 576 12.1 145.2
51-55 6 53 318 17.1 102.6
56-60 5 58 290 22.1 110.5
100 =  3590 982.6

x̄ =
∑ f .x 3590
= = 35 . 9
N 100
M . D x̄ =
∑ f |D x̄| = 982 . 6 = 9. 826
N 100

M . D x̄ 9.826
C . M . D x̄ = = = 0 .274
x̄ 35 .9

3.3.3 STANDARD DEVIATION

Standard deviation may be defined as the square root of the squared deviation may be defined as the
square root of the squared deviations measured form arithmetic average. It is also called the “Root-mean
square deviation.” Symbolically it is expressed by the small Greek letter S (Sigma). For calculating the
standard deviation the deviations are always taken from the mean, because, the sum of squared
deviations is minimum when form arithmetic mean measured. High value of standard deviations are
minimum when measured from arithmetic mean, high value of standard deviation denotes greater
variation & less uniformity and less measure of standard deviation denotes lesser variation and
greater uniformity. Whereas the lower value indicates that the averages are good representatives
of the data.
While the standard deviation is an absolute measure, the relative measure is known as coefficient of
variation.
Explained below are methods of calculating the standard deviation in individual, discrete and continuous
series.

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 18
I. (a) Individual Series: When deviations are measured from actual arithmetic mean.
i) Calculate the arithmetic mean of the given distribution
ii) Take the deviations from the actual mean (X-X).
iii) Square the individual deviations and find out their total.
Divide the squared deviations by the number of items and find out the square root to obtain
the S.D. Thus the formula

σ =
√ ∑ x2
N
 = Standard deviation
∑x 2
= sum of the squared deviations of “x” values measure from the actual arithmetic
mean
N = Number of observations
3.3.4 VARIANCE

The term variance was used by R.A Fisher in 1913, According to him “ it is the square of standard
(2). “ In the words of William Greenwald the variance is “The mean of the squared deviations
deviation (
about the mean of a series” Thus symbolically variance may be expressed as:

V =
∑ ( x − x̄ )2
N
Where, V = variance

∑ fd 12
( )
2

−
∑ fd 1
Variance = N N Variance and standard deviation are closely related each other, since,
the variance is the square of standard deviation and standard deviation is the square root of variance.

Example 1 Find the standard deviations and variance form the following.
14 15 17 22 23 25 24 16 18 26
Solution :Computation of standard deviation and coefficient of standard deviation.

x (x- x̄ ) (x- x̄ )2
14 -6 36
15 -5 25
17 -3 9
22 2 4
23 3 9

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 19
 25 5 25
24 4 16
16 -4 16
18 -2 4
26 6 36
x = 200  (x- x̄ )2 = 180

∑x
Arithmetic mean = x̄ = N
Here, x = 200, N = 10
Substituting the values in the formula,
200
= 20
x̄ = 10

σ =
N√ ∑ x2

σ =
10√√
180

 = 18 = 4.24
V= 18

Individual Series: When the deviations are taken from assumed mean
i. Take any value as the assumed mean and measure the deviations of ‘x’ from the assumed mean
and find its total.
ii. Square these deviations and find out their total.

√ ∑ dx 2
( )
∑ dx
2

σ = −
iii. Apply the formula N N
Where, ∑dx = sum of the deviations measured from the assumed mean
∑dx2 = sum of the squared deviations measured from the assumed mean
N = total number of deviations

Example 2 Find the standard deviation and coefficient of standard deviation from the following data
(x) : 458 459 450 448 457 462 439 454 451
Solution: Calculation of standard deviation and coefficient of standard deviation.

x dx dx2
A = 457
458 +1 1
459 +2 4
450 -7 49
448 -9 81
457 0 0
462 +5 25

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 20
 439 -18 324
454 -3 9
451 -6 36
∑x = 4078 ∑dx = -35 ∑dx2 = 529

∑x 4078
= 453 . 11
Arithmetic mean x̄ = N = 9

√ ∑ dx 2
( )
∑ dx
2

σ = −
N N
2
Here ∑dx = 529, ∑dx = -35, N=9

Substituting the values in the formula

√√ ( )
2
529 −35
σ = −
9 9
σ = 58.78 − 15.12
σ = √ 43.66
= 6.61
V = 43.66
Since the coefficient of standard deviation is low, the distribution is highly consistent.

II. Continuous Series

a) When deviations are calculated from actual arithmetic mean.
i) find the mid value (class marks) of the classes of the distribution
ii) calculate the arithmetic mean and take the deviation of mid values form the actual
mean.
iii) square the deviation and multiply them with their respective frequencies, and obtain their total.
iv) divide the sum of the products of frequencies and squared deviations by the sum of the
frequencies, and obtain the standard deviation.

Symbolically
σ =
√ ∑ fd 2
N
Σfd2 = Sum of the products of frequencies and squared deviations
Where, Σfd
N = Sum of the frequencies.

Example 3 Calculate standard deviation and coefficient of standard deviation from the following data.
Class f fx d d2 fd2
1–5 3 15 45 -10 100 1500

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 21
 6 – 10 8 12 96 -5 25 300
11 – 15 13 20 260 0 0 0
16 – 20 18 16 288 5 25 400
21 – 25 23 10 230 10 100 1000
26 – 30 28 2 56 15 225 450
75 975 3650

∑ fx
Arithmetic Mean = x̄ = N ∑f
∑fx = 975, N = 75
975
= 13
= 75

Standard Deviation =
σ =
√ ∑ fd 2
N Here ∑d2 = 3650, N = 75
Substituting the values in the formula

σ =
√ 3650
75
σ = √ 48.67
= 6.98
V= 48.67
As the coefficient of variation is high, the distribution is less uniform or more variable.

b) Continuous series: When deviations are measured form assumed mean.

i) Find the class-marks (mid-values) of classes of distribution.
ii) Take any value from the class-marks as the assumed mean and measure the deviations of mid-
values from the assumed mean.
iii) Multiply these deviations and multiply them with their respective frequencies and find out their
total.
iv) Square the deviations and multiply them with their respective frequencies and find out their total
v) Use the following formula

√ ∑ fdx 2
( ∑ fdx
)
2

= −
 N N

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 22
 Where Σfdx
Σfdx = Sum of the products of the deviations measured from assumed mean and their
respective frequencies.
Σfdx2 = Sum of the products of squared deviations measured from assumed mean and their
respective frequencies.
N = Sum of the frequencies.

Example 4.Calculates the standard deviation and its coefficient from the following data on age of
workers.
Age 25 - 29 30 - 34 35 - 39 40 - 44 45 - 49 50 - 54
Number of 8 40 76 44 20 6
Solution
persons
Calculating
the standard deviation and coefficient of standard deviation.

Age class Mid-value Number of workers fx dx fdx

fdx dx2 fdx2
(x) (f )
25 – 29 27 8 216 -15 -120 225 1800
30 – 34 32 40 1280 -10 -400 100 4000
35 – 39 37 76 2812 -5 -380 25 1900
40 – 44 42 44 1848 -0 0 0 0
45 – 49 47 20 940 5 100 25 500
50 – 54 52 6 312 10 60 100 600
194 7408 -740 8800

A +
∑ fdx
Arithmetic mean = x̄ = N here A = 42, Σfdx
Σfdx = -740, N = 194
Substituting the values in the formula
−740
42 +
x̄ = 194
= 42 + (-3.814)
= 38.186=

√ ∑ fdx 2
( ∑ fdx
) √
2

( ) = √ 45.361 − 14.550
2
8800 −740
= − = −
Standard deviation N N 194 194
= √30.811 = 5.55
Since the coefficient of standard deviation is low, the age of those workers is more consistencies or
fewer variables.

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 23
.

Statistics for Finance lecture note by Legese T. July 2009/2016 edtied Page 24