Explain VaR (Value at Risk)
Quant ResearchLevel: Intermediate
Explain VaR (Value at Risk)
Hints
- Specify the horizon, loss distribution and probability level.
- Consider tail behavior separately from scenarios absent in historical data.
Answer
For loss L, VaR at level α is inf{l:P(L≤l)≥α}, with horizon and sign convention specified. It is a quantile threshold, not the maximum possible loss and not the average loss beyond that threshold. Tail severity can change greatly without changing VaR, and VaR is not generally subadditive. Estimation depends on the distribution, data window and portfolio revaluation; backtest exceptions and their clustering without treating a passing test as proof of safety.
Source and licence: Quantitative Finance: From Mathematical Foundations to Trading Desk Reality — zcoderz · MIT · answer: Quantoros editorial