Topic 1-5 non-para estimation

Topic 1 Concept Checkers 错误题：2.3.5

---

Topic 2 concept check 错题：3.4.5

- Non-parametric estimation does not make restrictive assumptions about the underlying distributions, making it perfect if tail events are sparse.

- surrogate density function //连中线

- (1) age-weighted/hybrid approach: dacay parameter, reduce the impact of ghost effects and older events that may not reoccur

(2)volatility-weighted

(3)correlation-weighted(不用计算)

(4) 更复杂的filtered historical simulation: uses bootstrapping and combines the traditional historical simulation approach with rich volatility modeling.

non-para的优劣: 

---

TOPIC 3 backtesting VaR

concept check 错题：None

Highlights：

> Basel committee established 4 categories of causes for exceptions and guidance for supervisors:

(1) The basic integrity of the model is lacking;

(2) Model accuracy needs improvement;

(3) Intraday trading activity;

(4) Bad luck

---

Topic 4 VaR Mapping  - THREE THINGS: Positions, risk factors and factor exposures

notes错题 Nr. 5

Mapping advantage

(1) assist the risk manager in evaluating positions whose characteristics may change over time, s.a. fixed income securities

（固定收益衍生品经常变？）

(2)manage risk when a sufficient history of data for an investment does not exist, s.a. IPO

> Distinguish the position-based and return-based analysis.

(return-based analysis may fail to account for hidden risks and suffer from investments that have a short record of existence.)

> 啊 关于risk factor 和 risk exposure 有点难理解

>delta-normal VAR for linear combinations of basic component

---

TOPic 5

错题 1.3.5