NO.PZ2018122701000002
问题如下:
A risk analyst is comparing the use of parametric and non-parametric approaches for calculating VaR and is concerned about some of the characteristics present in the loss data. Which of the following distribution characteristics would make parametric approaches the favored method to use?
选项:
A.Skewness in the distribution
B.Fat tails in the distribution
C.Scarcity of high magnitude loss events
D.Heteroskedasticity in the distribution
解释:
C is correct.
考点non-parametric method
解析Non-parametric approaches can accommodate fat tails, skewness, and any other non-normal features that can cause problems for parametric approaches. However, if the data period that is used in estimation includes few losses or losses with low magnitude, non-parametric methods will often produce risk measures that are too low. Hence parametric methods would be more appropriate in those situations.
这个知识点掌握的不是很牢固,这道题还理解的不太透彻,想跟老师再学习一下:
参数法就是normal和lognormal两种,
AB:lognormal也是有偏或者肥尾啊?
C:选项是因为极端值少,所以用非参数法的优势就不大,但是如果参数法会不会更可能不合适啊?
D:异方差就导致不能很好估计参数,所以不能选。
谢谢老师