QTTM509

Tanima Dutta (PhD Economics)

Mittal School of Business
Lovely Professional University
 To reason out the difference and
complementarity between
measures of dispersion and
skewness, kurtosis
Lecture
Outcome
To interpret the results of different
measures of relative skewness

Tanima Dutta 2
What’s in the News?

Tanima Dutta 3
What’s in
The News?

Tanima Dutta 4
What’s in the
NEWS?

Tanima Dutta 5
What’s in the NEWS?

Tanima Dutta 6
What’s in the NEWS?

Tanima Dutta 7
What’s in the
NEWS?

Tanima Dutta 8
What’s in the
NEWS?

Tanima Dutta 9
Tanima Dutta 10
 Skewness

• Measure of skewness is the statistical

technique to indicate the direction and
extent of skewness in the distribution
of numerical values in the data set.

Tanima Dutta 11
Skewness

Tanima Dutta 12
What is the difference between
Skewness and Variation?

• Variation indicates the amount of spread or dispersion of

individual values in a data set around a central value, while
skewness indicates the direction of dispersion, that is, away
from symmetry.

• Variation is helpful in finding out the extent of variation

among individual values in a data set, while skewness gives
an understanding about the concentration of higher or lower
values around the mean value.

Tanima Dutta 13
Measures of Skewness

Absolute measure Relative measure

Tanima Dutta 14
Absolute Measure of Skewness

• For an asymmetrical distribution, the distance between mean and

mode may be used to measure the degree of skewness because the
mean is equal to mode in a symmetrical distribution.

Or

Tanima Dutta 15
Absolute Measure of Skewness

• then the distribution is positively skewed and vice-versa.

• The larger the difference between mean and mode, whether positive
or negative, more the asymmetrical distribution or skewness.

Tanima Dutta 16
Think

What happens when the difference between

mean and mode may be large in one
distribution and small in another, although
the shape of their frequency curves is the
same?

Tanima Dutta 17
Relative Measures of Skewness

Karl Pearson’s Bowley’s Kelly’s

coefficient of Coefficients of Coefficient of
skewness Skewness Skewness

Tanima Dutta 18
Karl Pearson’s Coefficient of Skewness

• Karl Pearson’s coefficient varies between ±3

Tanima Dutta 19
Example

• Data of rejected items during a production process is as follows:

No of rejects: 21–25 26–30 31–35 36–40 41–45 46–50 51–55 (per

operator)

No. of operators: 5 15 28 42 15 12 3

• Calculate the mean, standard deviation, and coefficient of skewness

and comment on the results
Tanima Dutta 20
Example

• The following is the information about the settlement of an industrial

dispute in a factory. Comment on the gains and losses from the point
of view of workers and that of management:

Tanima Dutta 21
Example

a. The sum of 50 observations is 500, its sum of squares is 6000 and

the median is 12. Find the coefficient of variation and coefficient of
skewness.

b. For a moderately skewed distribution, the arithmetic mean is 100,

the coefficient of variation is 35, and Pearson’s coefficient of
skewness is 0.2. Find the mode and the median.

Tanima Dutta 22
Bowley’s Coefficients of Skewness

Limit is ±1

Tanima Dutta 23
Think

• Why is the limit more in Karl Pearson’s coefficient and less in

Bowley’s coefficient?

Tanima Dutta 24
Example

• The data on the profits (in Rs lakh) earned by 60 companies is as

follows:

Profits: Below 10 10–20 20–30 30–40 40–50 50 and above

No. of Companies: 5 12 20 16 5 2

(a) Obtain the limits of profits of the central 50 per cent companies (b)
Calculate Bowley’s coefficient of skewness
Tanima Dutta 25
Example

Apply an appropriate measure of skewness to describe the following

frequency distribution.

Tanima Dutta 26
Example

• Compute the quartiles, median, and Bowley’s coefficient of skewness:

Tanima Dutta 27
Kurtosis

• Kurtosis refers to the degree of flatness or peakedness in the region

around the mode of a frequency curve.

Tanima Dutta 28
Types of Kurtosis

Tanima Dutta 29
Kurtosis via Moments

• According to R. A. Fisher, ‘A quantity of data which by its

mere bulk may be incapable of entering the mind is to be
replaced by relatively few quantities which shall
adequately represent the whole, or which, in other words,
shall contain as much as possible, ideally the whole, of the
relevant information contained in the original data.’

Tanima Dutta 30
Moments

• Moments represent a convenient and unifying method for

summarizing certain descriptive statistical measures.

• The measures of central tendency, variability and skewness describe a

frequency distribution, may be classified into two groups:

1. Percentile system

2. Moment system
Tanima Dutta 31
Moments

• The percentile system includes measures like median, quartiles,

decides, and percentiles.

• The value of these measures represents a given proportion of

frequency distribution.

Tanima Dutta 32
Moments

• The moment system includes measures like mean, average deviation,

and standard deviation.

• The term ‘moment’ is used in physics and refers to the measure of a

force which may generate rotation.

Tanima Dutta 33
Moments
The
possibility The amount of force needed and
of
generating
such a force
depends
upon The distance from the origin of the
point at which the force is applied.

Tanima Dutta 34
Moments

When we size of class intervals represents the

measure
moments ‘force’

deviation of the mid-value of each

class from an observation represents
the distance
Tanima Dutta 35
Hmmm

• Why are we studying

Kurtosis?

• How is it related to the

concept of “Moments” which
originated in Physics?
Tanima Dutta 36
Moments

• While calculating moments, if deviations are taken from the actual

mean, then such moments are denoted by the Greek letter µ (mu).

• On the other hand, if deviations are taken from some assumed mean
(or arbitrary value other than zero), then moments are denoted by
Greek letter ν (nu) or µ′.

Tanima Dutta 37
Kurtosis Via
Moments
Remember why we are
studying Moments?

Of course to reach
Kurtosis

Tanima Dutta 38
How to measure Kurtosis?

• The fourth moment about the mean is a measure of Kurtosis (or

flatness) of the frequency curve.

4
← Kurtosis

Tanima Dutta 39
How to measure Kurtosis?

• The first moment µ1, about origin gives the mean and is a measure of
central tendency.

← A.M.

• The second moment µ2 about the mean is known as variance and is a

measure of dispersion.

← Variance
Tanima Dutta 40
How to measure Kurtosis?

• The third moment µ3, about the mean indicates the symmetry or
asymmetry of the distribution; its value is zero for symmetrical
distribution

← Skewness

Tanima Dutta 41
Tanima Dutta

• Why are you confusing me with so many

mathematical notations and moments when I
want to find out kurtosis?

• Tell me how to say whether a distribution is

platykurtic or leptokurtic.

42
Here is the answer

• For a normal distribution so that and hence any distribution having

will be peaked more sharply than the normal curve known as
leptokurtic (narrow) while if , the distribution is termed as
platykurtic (broad).

Tanima Dutta 43
Moments in Standard Units

• When expressed in standard units moments about the mean of the

population are usually denoted by the Greek letter (alpha).

• These notations are given by Karl Pearson

Tanima Dutta 44
Moments in Standard Units

• These notations are given by R A Fisher

Tanima Dutta 45
Example

• Find the standard deviation and kurtosis of the following set of data
pertaining to kilowatt hours (kwh) of electricity consumed by 100
persons in a city.

Consumption (in kwh): 0–10 10–20 20–30 30–40 40–50

Number of users: 10 20 40 20 10

Tanima Dutta 46
Example

• Calculate the value of γ1 and γ2 from the following data and interpret
them.

Profit (Rs in lakh): 10–20 20–30 30–40 40–50 50–60

Number of companies: 18 20 30 22 10

• Comment on the skewness and kurtosis of the distribution

Tanima Dutta 47