We provide a framework based on the unbiased extreme value volatility estimator to predict long and short position value-at-risk (VaR). The given framework incorporates the impact of asymmetry, structural breaks and fat tails in volatility. We generate forecasts of long and short position VaR for the cases when future structural breaks are known as well as unknown. We evaluate its VaR forecasting performance using various backtesting approaches for both long and short positions and compare the results with that from return based models. Our findings indicate that incorporating the impact of structural breaks in volatility indeed improves the accuracy of VaR forecasts of the proposed framework.

中文翻译：

---

使用无偏极值波动率估算器的结构断裂风险值

我们提供基于无偏极值波动率估计量的框架来预测多头和空头头寸风险值（VaR）。给定的框架包含了不对称性，结构断裂和波动中的肥尾现象的影响。当已知和未知未来结构性断裂时，我们会生成多头和空头头寸VaR的预测。我们使用各种多头和空头回测方法评估其VaR预测性能，并将结果与​​基于收益的模型的结果进行比较。我们的发现表明，将结构性突破的影响纳入波动确实会提高所提议框架的VaR预测的准确性。