Predicting extreme value at risk: Nonparametric quantile regression with refinements from extreme value theory
Predicting extreme value at risk: Nonparametric quantile regression with refinements from extreme value theory
复制标题
DOI：
10.1016/j.csda.2012.03.016
复制
发表时间：
2012-12
期刊：
影响因子：
--
通讯作者：
Julia Schaumburg
中科院分区：
文献类型：
--
作者：
Julia Schaumburg
引入了一个框架，允许我们将非参数分位数回归应用于在任何感兴趣的概率水平下的风险价值(VaR)预测。利用单调双核局部线性估计估计指数收益分布的中等(1%)条件分位数。对于极值(0.1%)分位数，非参数分位数回归与极值理论相结合。在VaR预测研究中，利用经验数据和模拟数据考察了所提出的估计器捕捉市场风险的能力。可能由于其灵活性，新模型的样本外预测性能被证明优于竞争模型。
A framework is introduced allowing us to apply nonparametric quantile regression to Value at Risk (VaR) prediction at any probability level of interest. A monotonized double kernel local linear estimator is used to estimate moderate (1%) conditional quantiles of index return distributions. For extreme (0.1%) quantiles, nonparametric quantile regression is combined with extreme value theory. The abilities of the proposed estimators to capture market risk are investigated in a VaR prediction study with empirical and simulated data. Possibly due to its flexibility, the out-of-sample forecasting performance of the new model turns out to be superior to competing models.