Journal of Banking and Finance 72 (2016) 121–132

VaR预测的评价：一组新的多变量后验分析

作者：Dominik Wied (Universit?t zu K?ln; Technische Universit?t Dortmund), Gregor N.F. Wei? (Universit?t Leipzig; Technische Universit?t Dortmund), Daniel Ziggel (FOM Hochschule für Oekonomie & Management)

摘要：为了识别多变量VaR预测聚类，我们开展了两项新的测试。第一，我们通过CUSUM-tests来探究在VaR-violation矩阵中的非稳定预期。第二，我们采用了χ2检验，以探测VaR预测中的横截面数据和时间序列数据的可靠性。更进一步的，我们将新的后验分析和非条件覆盖检验结合，从而得到两种新的多变量条件覆盖检验的后验分析。我们新的后验分析控制了一系列银行业务线相关的组合风险，仿真研究的结论证明这种后验分析是非常有效的。我们实证研究结果还告诉了监管者如何利用多变量后验分析来监测银行体系。

关键词：模型风险，多变量后验分析，VaR，系统风险

Evaluating Value-at-Risk forecasts: A new set of multivariate backtests

Dominik Wied (Universit?t zu K?ln; Technische Universit?t Dortmund), Gregor N.F. Wei? (Universit?t Leipzig; Technische Universit?t Dortmund), Daniel Ziggel (FOM Hochschule für Oekonomie & Management)

ABSTRACT

We propose two new tests for detecting clustering in multivariate Value-at-Risk (VaR) forecasts. First, we consider CUSUM-tests to detect non-constant expectations in the matrix of VaR-violations. Second, we propose χ2 -tests for detecting cross-sectional and serial dependence in the VaR-forecasts. Moreover, we combine our new backtests with a test of unconditional coverage to yield two new back- tests of multivariate conditional coverage. Results from a simulation study underline the usefulness of our new backtests for controlling portfolio risks across a bank’s business lines. In an empirical study, we show how our multivariate backtests can be employed by regulators to backtest a banking system.

Keywords: Model risk, Multivariate backtesting, Value-at-Risk, Systemic risk

原文链接：http://dx.doi.org/10.1016/j.jbankfin.2016.07.014

翻译：郎彪