Module: | MODULE B: RISK MANAGEMENT
Q351: Consider the following statements regarding the Historical Simulation method for computing Value at Risk (VaR):
1. The method relies heavily on the fundamental mathematical assumption that asset returns always follow a perfectly normal distribution.
2. It uses actual historical daily market price movements applied to the current portfolio to generate a simulated distribution of future returns.
3. A major limitation of this method is its inability to account for unprecedented structural market shifts or extreme events that have not occurred during the historical lookback period.
2. It uses actual historical daily market price movements applied to the current portfolio to generate a simulated distribution of future returns.
3. A major limitation of this method is its inability to account for unprecedented structural market shifts or extreme events that have not occurred during the historical lookback period.
✅ Correct Answer: B
The correct answer is B. Statement 2 is correct: The Historical Simulation method calculates VaR by taking the current portfolio and revaluing it using the actual historical daily price changes observed over a specific lookback period (e.g., the last 250 to 500 trading days). Statement 3 is correct: Its primary limitation is the "history repeats itself" fallacy.
If a black swan event or a structural regime shift has not occurred within the historical window analyzed, the model will assign a zero probability to that risk, potentially undercapitalizing the bank.
Statement 1 is incorrect: Historical Simulation is explicitly a "non-parametric" method, meaning it does NOT assume asset returns follow a normal distribution.
It relies purely on the empirical, actual distribution of past data, fat tails and all.
The Variance-Covariance (Parametric) method is the one that assumes a normal distribution.
If a black swan event or a structural regime shift has not occurred within the historical window analyzed, the model will assign a zero probability to that risk, potentially undercapitalizing the bank.
Statement 1 is incorrect: Historical Simulation is explicitly a "non-parametric" method, meaning it does NOT assume asset returns follow a normal distribution.
It relies purely on the empirical, actual distribution of past data, fat tails and all.
The Variance-Covariance (Parametric) method is the one that assumes a normal distribution.