返回列表 发帖

真诚请教FRM handbook第六版关于一道Kurtosis的问题

Handbook 第六版

Example 2.1, FRM exam 2009 -question 2-3

   An analyst gathered the following information about the return distributions for two portfolios during the same time period:


Portfolio
Skewness
Kurtosis

A
-1.6
1.9

B
0.8
3.2



The analyst states that the distribution for portfolio A is more peaked than a normal distribution and that the distribution for Portfolio B has a long tail on the left side of the distribution. which of the following is correct?

a. The analyst's assessment is correct
b. The analyst's assessment is correct for Portfolio A and incorrect for B
c. The analyst's assessment is not correct for A but correct for B
d. The analyst's assessment is incorrect for both portfolios.

The answer is b. A has less kurtosis than normal distribution, which implies that it is more peaked.

这个对吗?不是Kurtosis<3, less peaked?

还有32页的Figure2.4也看着糊涂, 不是高峰胖尾,为什么反过来了? (这个图在第五板的36页也有)

Kurtosis=3 is normal dist., >3 is more peak than a normal, <3 is less peak than a normal....

TOP

so the ans is right.

TOP

答案b说的跟你说的是反的啊?
2# mjsnoopy

TOP

很難,我也不太明白

TOP

Also, if you read the figure that I mentioned in post(pg32 in 6ed and pg36 in 5ed of handbook), you will find it shows-
Kurtosis <3 more peaked than normal distribution;
Kurtosis >3, less peaked.

This is what I really confused!

TOP

就没人知道吗?

TOP

答案是b
Skewness        只用比较positive or negative sign 就可以了positive 就是skew to the right, long right hand tail(not fat tail)
Kurtosis 是用比较是比3大还是比3小, 数字越大,表示越flat,fat tail。
二楼的回答是错的

TOP

还有,对比nornal distribution,高峰胖尾同时出现是不可能的。应该是你理解错了吧

TOP

9# johnluo721
老实说我不是很理解这个handbook的说法(也想弄清楚)。

至少金程的书是这么说“高峰胖尾”。很多教材也是这么画的图。

比方说啊: PRM  handbook(v2)-Math foundation of risk management p57说

"If the data were normally distributed, the momemt of coefficient of kurtosis would be 3.0; The computed index value is greater than 3, indicating that the return data in question are more peaked than normal distribution and therefore are leptokurtic. ...."

跟你说的相反,跟FRM handbook上说的也相反。

TOP

返回列表