P2.T5.209. Value at risk (I. Historical simulation)

Suzanne Evans

Well-Known Member
Questions:

209.1. You colleague John proposes the use of a simple historical simulation (HS) approach (i.e., without semi-parametric adjustment) to compute the value-at-risk (VaR) of a portfolio, based on the last 250 trading days. Which of the following, if true, would be most problematic and most implicate the HS approach as inappropriate or dubious to the task?

a. The portfolio returns are skewed and heavy-tailed
b. Although the portfolio does not exhibit serial correlation over time, the assets within the portfolio exhibit significant correlation in returns on a given day
c. The bank's internal risk system requires a 99.99% VaR (i.e., extreme tail confidence) with significant precision
d. The portfolio contains options and derivatives which exhibit non-linearities

209.2. Over the last 300 trading days, the five worst daily losses were: -21.0 million, -19.0 million, -14.0 million, -13.0 million, and -9.0 million. If the historical window is these 300 daily profit & loss (P&L) observations, what is the DIFFERENCE between the 99.0% daily expected shortfall (ES) and 99.0% daily HS VaR?

a. $3.7 million
b. $5.0 million
c. $6.8 million
d. $8.3 million

209.3. Which of the following VaR approaches is most likely to produce a VaR estimate that exceeds the maximum loss in the historical data set?
a. Simple historical simulation (HS)
b. Bootstrapped historical simluation
c. Age-weighed historical simulation
d. Volatility-weighted historical simulation

Answers:
 
Top