Scribd
Upload
16 views46 pages

Moments, Skewness & Kurtosis

Uploaded by

mambayar.mambu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
16 views46 pages

Moments, Skewness & Kurtosis

Uploaded by

mambayar.mambu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 46

Moments, Skewness & Kurtosis

Moments
The word 'Moment' is a mechanical term, which in Physics, refers to a measure
of force. But in statistics, it may be defined as "the arithmetic average of
certain power of the deviations of the items from their arithmetic mean".
Accordingly, there can be Moments of various orders viz. first order, second
order, third order, fourth order and so on.
The Moment of the first order, or the First Moment is the arithmetic
mean of the first power of the deviations of the items from their
arithmetic mean. It is represented by u₁ (mu one) as follows:
Moments about an arbitrary value.
Such Moments are otherwise called assumed Moments. These are computed
as arithmetic mean of certain power of deviations of items from an arbitrary
value. These Moments are symbolically represented as follows.
SKEWNESS
A distribution is said to be 'skewed' when the mean
and the median fall at different points in the
distribution, and the balance (or centre of gravity) is
shifted to one side or the other-to left or right.

Measures of skewness tell us the direction and the


extent of Skewness. In symmetrical distribution the
mean, median and mode are identical. The more the
mean moves away from the mode, the larger the
asymmetry or skewness
Characteristics of Symmetrical
Series
1. The values of the series when plotted on a graph paper give a bell shaped curve as
under.

2. The values of Mean, Median and Mode are identical. (MEAN = MEDIAN = MODE)
3. Both the extreme quartiles i.e. Q3 and Q₁ are equidistant from the Median i.e. (Q3-
M)=(M-Q1).
4. Both the extreme Percentiles, and Deciles are equidistant from theMedian.
(P90-M)=(M-P10), and
(D9-M)=(M-D1)
5. The sum of the positive deviations from Median is equal to the sum of the
negative deviations from Median i.e. Σ (X-M) = 0
6. The frequencies are equally distributed at the points of equal deviations
from Mode.
7. ẞ1 (Beta one) is equal to zero.
8. 2 (Beta two) is equal to 3.
9. Y1 (Gamma one) is equal to zero.
Tests of Skewness

A series in which the above characteristics are absent is called a skewed, or an


asymmetric series. Thus, the following would be the tests of skewness:
(i) The values of Mean, Median and Mode are different.i.e.X ≠ M ≠ Z.
(ii) The difference between the upper Quartile, and the Median is not equal to the
difference between the Median and the lower Quartile, i.e. (Q3-M) ≠ (M-Q1)
(iii) The extreme values of the Percentiles and Deciles are not equidistant from Median
i.e. (P90-M) ≠ (M-P10)
(D9-M) ≠ (M-D₁).
(iv) The sum of the positive deviations from the Median is not equal to the
sum of the negative deviations from the Median i.e. Σ (X-M) ≠ 0
(v) The frequencies are not equally distributed at the points of equal
deviationfrom Mode.
(vi) ẞ1 (Beta one) is not equal to zero……..i.e.,ẞ₁ ≠ 0
(vii) Y1 (gamma one) is not equal to zero…...i.e.Y1 ≠ 0
(viii) 2 (Beta two) is not equal to 3……..i.e.,B2 ≠ 3
(ix) The values of the variables when plotteda bell shaped curve.
Types of Skewness
Broadly speaking, there are two types of skewness: They are (1) Positive
skewness and (2) Negative skewnes.
• Positive skewness
A series is said to have positive skewness when the following characteristics
are noticed :
(i) Mean > Median > Mode.
(ii) The right tail of the curve is longer than its left tail, when the data are
plotted through a histogram, or a frequency polygon.
(iii) The formula of Skewness and its coefficient give positive figures.
• Negative skewness
A series is said to have negative skewness, when the following characteristics are
noticed :
(i) Mode > Median > Mean.
(ii) The left tail of the curve is longer than the right tail, when the data are plotted
through a histogram, or a frequency polygon.
(iii) The formula of skewness and its coefficient give negative figures.

Thus, a statistical dostribution may be of three types viz.


(ii) Symmetric
(ii) Positively skewed
(iii) Negatively skewed.

The following examples would show the above three types of distribution, and their
respective characteristics.
Positively skewed Negatively skewed.
Graphical Measures of Skewness
Different measures of Skewness
Like the measures of dispersion, Skewness can be computed in two ways viz. absolute
and relative. The relative skewness is otherwise called coefficient of skewness. For a
comparative study of more than one distribution, coefficientof skewness is found
out.However, there are different measures of skewness propounded by different authors
on the basis of the different tests of skewness as under:
1. General measures of Skewness
(a) Absolute skewness
This is obtained by finding the difference between any two measures of dispersion viz :
Mean, Median
SK = and Mode. Thus, Skewness
X-M or or X-Z or M-Z

Any positive value obtained by any of the above formulae is marked as the extent of the
positive skewness. Any negative value obtaimed by any of the above formulae is marked
as the extent of the negative skewness of the distribution. If the result produced is zero,
it signifies the absence of skewness in the distribution.
(b) Co-efficient of skewness
This is obtained by dividing the Skewness by any measure of dispersion.
Thus,Coeff. Sk= or or or
2. Relative Measures of Skewness

There are four measures of skewness:


ẞ and y Coefficient of skewness
Karl Pearson's Coefficient of skewness
Bowley's Coefficient of skewness
Kelly's Coefficient of skewness
ẞ and y Coefficient of Skewness
Kerl Pearson’s of
Skewness…..01
Kerl Pearson’s of
Skewness…..02
Example.
Bowley’s Coefficient of
Skewness…..01
Bowley’s Coefficient of
Skewness…..02
Kelly's Coefficient of
skewness….01
Kelly’s Coefficient of Skewness….02
KURTOSIS
(i) Meaning and Purpose
Kurtosis is a statistical measure which is calculated on the basis of
Moments to bring about another characteriftic of a series. This
characteristic refers to the flatness, or peakedness of the curve obtained
from a series. A series, when plotted on a graph paper may produce either
of the following three types of curves:
1. Symmetric, or normal curve which is neither more flat nor more
peaked.
2. Asymmetric curve which is more flat than a normal one.
3. Asymmetric curve which is more peaked than a normal one.
Thus, through the measure of Kurtosis we can know, if a distribution would
give us a normal curve, a curve which is more flat than a normal one, or a
curve which is more peaked than a normal curve.

It is to be noted that the word Kurtosis is a Greek word which means


bulginess. Similarly, in Greek language, a normal curve is called
Mesokurtic, a more flatted curve is called Platykurtic, and a more peaked
curve is called a Leptokurtic.
These three types of curves are
given under.
Different Methods of Kurtosis
• There are two constants or co-efficients through which Kurtosis of a series
is determined. They are (1) ẞ2 and (2) Y2.
• 1. ẞ2 (Beta two)
According to Prof. Karl Pearson, Kurtosis of a series can be measured by ẞ2
(beta two) which is equal to.
Indications
If ẞ2 = 3, it will give a normal or a Meso kurtic curve.
If ẞ2 > 3, it will give a Lepto kurtic, or a more peaked curve.
Ifẞ2 <3, it will give a platy kurtic, or a more flat-topped curve.
2. y2 (Gamma two)
According to R.A. Fisher, kurtosis is measured by y2 which is equal to ẞ2-3.
Indications
If y2 =0, it will give a meso kurtic or a normal curve.
If y2 > 0, it will give a more peaked or a lepto kurtic curve.
If y2 < 0, it will give a more flat-topped or a platy kurtic curve.
EXAMPLE
• From the following frequency distributions, indicate the type of
Kurtosis both by (i) algebraic and (i) graphic method.

You might also like