Explanation:
A 99% VaR implies that losses should exceed the VaR estimate approximately 1% of the time. An average exceedance rate of 0.4%, which is lower than the expected 1%, indicates that losses are exceeding the VaR less frequently than predicted. This suggests that the VaR estimates are generally higher than necessary, indicating overestimation of risk or conservativism.
A is incorrect. If the models were underestimating risk, the exceedance rate would be higher than 1%.
B is incorrect. An accurate model would have an exceedance rate close to 1%.
D is incorrect. The exceedance rate is a direct measure of how often actual losses exceed the predicted VaR, thus directly informing us about the model's conservativeness.
Things to Remember
Ultimate access to all questions.
Q.6476 A study examining the backtesting results of 20 bank holding companies found an average exceedance rate of 0.4% for their one-day 99% VaR models. What does this average exceedance rate suggest about the overall conservatism of these models?
A
The models are, on average, underestimating risk.
B
The models are, on average, accurately capturing risk.
C
The models are, on average, overestimating risk.
D
Further information is needed to reach a conclusion.
No comments yet.