Paper: arXiv 2402.07134

Authors: Cathy W. S. Chen, Takaaki Koike, Wei-Hsuan Shau

Abstract

This research incorporates realized volatility and overnight information into risk models, wherein the overnight return often contributes significantly to the total return volatility. Extending a semi-parametric regression model based on asymmetric Laplace distribution, we propose a family of RES-CAViaR-oc models by adding overnight return and realized measures as a nowcasting technique for simultaneously forecasting Value-at-Risk (VaR) and expected shortfall (ES). We utilize Bayesian methods to estimate unknown parameters and forecast VaR and ES jointly for the proposed model family. We also conduct extensive backtests based on joint elicitability of the pair of VaR and ES during the out-of sample period. Our empirical study on four international stock indices confirms that overnight return and realized volatility are vital in tail risk forecasting.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 7.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a solid extension to the CAViaR model family by integrating overnight returns and realized measures using Bayesian estimation and an asymmetric Laplace distribution. It demonstrates strong empirical rigor through extensive backtesting on international stock indices using joint elicitability framework for VaR and ES.

Research Flowchart

  flowchart TD
    A[Goal: Enhance VaR & ES Forecasting<br>using Overnight & Realized Info] --> B[Data Inputs: 4 International Stock Indices<br>Intraday & Overnight Returns]
    B --> C[Model Development: Propose RES-CAViaR-oc<br>Semi-parametric Asymmetric Laplace Framework]
    C --> D[Estimation & Forecasting:<br>Bayesian Inference for Joint VaR/ES Nowcasting]
    D --> E[Evaluation: Out-of-Sample Backtesting<br>via Joint Elicitability]
    E --> F[Findings: Overnight Return & Realized Volatility<br>Significantly Improve Tail Risk Models]