1
KURTOSIS
2 AGENDA
• Introduction
• Importance of Kurtosis
• Types of Kurtosis
• Measures of Kurtosis
• Example
• Advantages
• Limitations and Applications
• Conclusion
3 INTRODUCTION
• Word Kurtosis comes from Greek word 'Kurtos' meaning humped.
• Gives an idea about shape of a frequency distribution.
• It is the measure of peak of distribution and indicates how high the
distribution is around the mean. [1]
• Indicates whether a frequency distribution is a flat, normal, or peaked shape.
4 IMPORTANCE OF KURTOSIS
• Used as measure to define risk an investment carries.
• High kurtosis of return distribution implies that an investment will yield occasional
extreme returns
• Nature of the investment to generate higher returns can also be predicted from the
value of the calculated kurtosis.
• Greater the excess for any investment data set, greater will be its deviation from the
mean.
5 TYPES OF KURTOSIS
• Karl Pearson classified curves into three types on the basis of shape of their
peaks. These are:-
•Leptokurtic
•Mesokurtic
•Platykurtic
6 LEPTOKURTIC
• When peak of a curve becomes relatively high then that curve is called
Leptokurtic.
•Too much concentration of items near the centre.
•Leptokurtic has a kurtosis that is + .
7 PLATYKURTIC
• When the curve is flat-topped, then it is called Platykurtic.
•There is less concentration of items near the centre.
•Platykurtic has a kurtosis that is - .
8 MESOKURTIC
• Since normal curve is neither very peaked nor very flat topped, so it is taken as a basis for
comparison.
• This normal curve is called Mesokurtic.
•There is equal distribution around the centre value.
•In such case mean, median, and mode are equal.
•Mesokurtic has a kurtosis of 0.
9 MEASURE OF KURTOSIS
• There are two measure of Kurtosis:
Karl Pearson’s Measures of Kurtosis
Kelly’s Measure of Kurtosis
10 KARL PEARSON’S MEASURES OF KURTOSIS
• For calculating the kurtosis, second and fourth central moments of variable
are used. For this, formula given by Karl Pearson is used:
β 2= or
where,
= Second order central moment of distribution
= Fourth order central moment of distribution
11 INTERPRETATION
• If = 3 or = 0
𝑇ℎ𝑒𝑛 𝑡ℎ𝑒 𝑐𝑢𝑟𝑣𝑒 𝑖𝑠 𝑠𝑎𝑖𝑑 𝑡𝑜 𝑏𝑒 𝑚𝑒𝑠𝑜𝑘𝑢𝑟𝑡𝑖𝑐.
• If < 3 or = 0
𝑇ℎ𝑒𝑛 𝑡ℎ𝑒 𝑐𝑢𝑟𝑣𝑒 𝑖𝑠 𝑠𝑎𝑖𝑑 𝑡𝑜 𝑏𝑒 platy𝑘𝑢𝑟𝑡𝑖𝑐.
• If > 3 or = 0
𝑇ℎ𝑒𝑛 𝑡ℎ𝑒 𝑐𝑢𝑟𝑣𝑒 𝑖𝑠 𝑠𝑎𝑖𝑑 𝑡𝑜 𝑏𝑒 lepto𝑘𝑢𝑟𝑡𝑖𝑐.
12 KELLY’S MEASURE OF KURTOSIS
• Kelly has given a measure of kurtosis based on percentiles. The formula is
given by:
β2 =
where,
P75, P25, P90, P10 are 75th, 25th , 90th and 10th percentiles of dispersion
respectively.
13 INTERPRETATION
• If > 0.26315
then the distribution is platykurtic.
• If < 0.26315
then the distribution is leptokurtic.
14 EXAMPLE
Ques: First four moments about mean of a distribution are 0, 2.5, 0.7 and 18.75. Find coefficient of
kurtosis.
Soln:
We have,
Kurtosis is β2= = = = 3.
As = 3, the curve is mesokurtic.
15 ADVANTAGES
• Mostly kurtosis will be in positive form and advantage is that distribution about
the means get tighter as the mean gets larger.
• When excess kurtosis in flat, it means probability of generating a high return from
investment is low and will generate high returns in only a few scenarios.
• High excess kurtosis means that result can swing both ways. It means generated
results can either be very high or very low.
16 LIMITATIONS AND APPLICATIONS
• Disadvantage is that it will not have negative or undefined form.
• Does not measure anything about peak.
Applications:
Parameter of kurtosis usually employed in financial analysis.
Useful to measure, whether there is a problem with outliers in a data set.
Larger Kurtosis – Outlier problem
Soln: Choose alternative Statistical methods.
Rarely used in other fields.
17 CONCLUSION
•Kurtosis indicates whether a frequency distribution is a flat, normal, or picked shape.
•Kurtosis is a measure of the peakedness of a distribution.
•The larger value of kurtosis, the more peaked will be the distribution.
•The kurtosis is calculated and reported either as an absolute or a relative value.
•Absolute kurtosis is always a positive number.
•Absolute kurtosis of a normal distribution is 3.
18
THANK YOU