KURTOSIS
Kurtosis-
definition, explanation and relevance to finance
Kurtosis in stats is used
to describe the distribution of the data set. Kurtosis depicts to what extent
the data set points of a particular distribution differs from the data of a
normal distribution. Kurtosis is used to determine whether a distribution
contains extreme values. In the area of finance the kurtosis is used to measure
the volume of financial risk associated with any instrument or a transaction.
More the kurtosis more is the financial risk associated with the concerned data
set.
Skewness is a measure of
symmetry in a distribution whereas the kurtosis is the measure of heaviness or
the density of distribution tails. These two factor differ from each other in
their definition, Kurtosis is an important descriptive statistic of data
distribution. An excess kurtosis is a metric which compares distribution
kurtosis against normal distribution kurtosis.
Excess Kurtosis= Kurtosis-3
Below is the pictorial representation
of the kurtosis (all three types, each one is explained in detail in the
subsequent paragraph)
Graphical
representation of kurtosis in simple terms
Types
of Kurtosis: There are three types of kurtosis
1.
Mesokurtic:
If the kurtosis of data falls close to zero or equal to zero, it is referred to
as Mesokurtic. This means that the data set follows a normal distribution. The series2
line in the above picture represents a Mesokurtic distribution. In finance such
a pattern depicts risk at a moderate level.
2.
Leptokurtic:
When kurtosis is positive on in other terms more than zero, the data falls
under leptokurtic kurtosis. Leptokurtic kurtosis has heavy steap curves on both
the sides indicating the heavy population of outliers in the data set. In terms
of finance a leptokurtic distribution shows that the return on investment may be
highly volatile on huge scale on either sides. An investment following
leptokurtic distribution is said to be a risky investment but it can also
generate hefty returns to compensate for the risk. The series 3 curve on the
above picture represents the leptokurtic distribution.
3.
Platykurtic:
whenever the kurtosis is less than zero or negative, it refers to platykurtic
Kurtosis. The distribution set follows subtle or pale curve and those curve
indicates the small number of outliers in a distribution. An investment falling
under platykurtic are usually demanded by investors because of small
probability to generate extreme return. Also the small outliers and flat tail
indicates the less risk involved in such investments. The series 1 in the above
graphical representation depicts a platykurtic distribution or a safe
investment.
Significance of kurtosis in finance
and its application for the investment world:
From
the perspective of investors, high kurtosis of the return distribution implies
that an investment will yield occasional extreme return. This can swing both
the ways that is either positive returns of extreme negative returns. Thus such
an investment carried high risk. Such a phenomenon is known as kurtosis risk.
The skewness measures the combined size of the two tails, the kurtosis measure
the distribution among the values in these tails.
When
the kurtosis distribution is calculated on any data set of a particular
investment, the risk of the investment against the probability of generating
returns. Depending on the value of kurtosis and the type of kurtosis it belongs
to, the investment predictions can be made by the investment advisors. Based on
the predictions advisors will advise the strategy and investment agenda to the
investor and they will chose to go about the investment. To calculate kurtosis
in excel, there is a built in function Kurt in excel.
Difference
between kurtosis and Skewness in statistics and finance:
· Skewness
is a measure of the degree of lopsided nature (with one side of the curve lower
than the other) in the frequency distribution of a dataset. Whereas kurtosis is
a measure of deviation in the normal distribution
· Skewness
is indicator of lack of symmetry in the distribution whereas kurtosis is a
measure of pointedness (sharpness of the curve) of the peak in the frequency
distribution.
· Skewness
is an indicator of lack of equivalence in the frequency distribution, on the
other hand kurtosis is the measure of data which is peaked or flat in relation
with the normal distribution.
· Skewness
represents the amount and direction of the skew whereas the kurtosis represents
the length and sharpness of the peak compared with the normal distribution.
· Kurtosis
show how tall the deviation is from the central peak.
Advantages/uses of kurtosis
When
the kurtosis is calculated on the data set of the investment, the value
obtained can be used to depict the nature of the investment. Greater the
deviation from the mean means the returns are also high for that particular investment.
When the excess kurtosis in flat, it means the probability of generating high
return from the investment is low and will generate high returns in only few
scenarios, on a regular basis the return is not so high on the investment. High
excess kurtosis means that the return on the investment can swing both the
ways. It means the generated returns can either be very high or very low as per
the outliers in the distribution. When the kurtosis is negative, it indicates
that the deviation of data set from the mean is flat.
Conclusion.
To
sum it up, kurtosis is used as a measure to define the risk an investment
carries. The nature of investment to generate higher returns can also be
predicted from the value of the calculated kurtosis. Greater the excess
kurtosis for any investment data set, greater will be its deviation from mean.
This mean such a investment has potential to generate higher returns or to
deplete the investment value to greater extent. Excess kurtosis closer to zero
or a flat deviation from the mean depicts that the investment will have lesser
probability to generate high returns. The kurtosis can be used to define the
financial risk of the investment. For investment advisor kurtosis is a crucial
factor to define the investment risk associated with the portfolio or the fund.

Comments
Post a Comment