Вы находитесь на странице: 1из 8

# Measuring Skewness

and Kurtosis

 CMRA
Why are Returns Skewed?

## Portfolio containing Portfolio containing long

short put options call options

 CMRA 2
Defining Skewness
Skewness is the standardized 3rd central
moment of a distribution

E ( X − µ )3
s =
σ 3

## ♦ Positive skewness indicates a long right tail

♦ Negative skewness indicates a long left tail
♦ Zero skewness indicates a symmetry around the mean

 CMRA 3
Calculating Skewness
Given a set of returns rt, t = 1,2,…T

1 T rt − r 3
sˆ = ∑ ( )
T t =1 σˆ

## Where r¯ and σ^ are the estimated average and standard deviation

1 T 1 T
r = ∑ rt σˆ = ∑ ( r − r ) 2

T − 1 t =1
t
T t =1

 CMRA 4
♦ A gamma distribution is a better proxy for skewed
portfolios
# of σ’s necessary
Skewness to achieve 99%
-2.83 3.99

-2.00 3.61
-1.00 3.03
-0.67 2.80

## 0.00 2.33 Symmetric (Normal Distribution)

0.67 1.83
1.00 1.59
2.00 0.99
2.83 0.71

 CMRA 5
Why are Returns Kurtotic?

## ♦ Frequent medium to large changes ♦ Very frequent small changes

♦ Less frequent very large changes ♦ Less frequent very large changes

## Positive excess kurtosis Negative excess kurtosis

(internet stocks) (Pegged exchange rate)

 CMRA 6
Defining Kurtosis
Kurtosis is the standardized 4th central moment of a
distribution

E ( X − µ )4
κ =
σ 42
excess_κ = κ − 3

## ♦ The kurtosis for the normal distribution is 3

♦ Positive excess kurtosis indicates flatness (long, fat tails)
♦ Negative excess kurtosis indicates peakedness

 CMRA 7
Calculating Kurtosis
Observatio n of a set of returns : rt , t = 1, 2 ,  T .

1 T rt − r 4
κˆ = ∑( )
T t =1 σˆ

1T
r = ∑rt
T t =1
1 T
σˆ = ∑
T −1 t =1
(rt − r )2

 CMRA 8