Paper: arXiv 2402.09985

Authors: Rangika Peiris, Chao Wang, Richard Gerlach, Minh-Ngoc Tran

Abstract

A semi-parametric joint Value-at-Risk (VaR) and Expected Shortfall (ES) forecasting framework employing multiple realized measures is developed. The proposed framework extends the realized exponential GARCH model to be semi-parametrically estimated, via a joint loss function, whilst extending existing quantile time series models to incorporate multiple realized measures. A quasi-likelihood is built, employing the asymmetric Laplace distribution that is directly linked to a joint loss function, which enables Bayesian inference for the proposed model. An adaptive Markov Chain Monte Carlo method is used for the model estimation. The empirical section evaluates the performance of the proposed framework with six stock markets from January 2000 to June 2022, covering the period of COVID-19. Three realized measures, including 5- minute realized variance, bi-power variation, and realized kernel, are incorporated and evaluated in the proposed framework. One-step-ahead VaR and ES forecasting results of the proposed model are compared to a range of parametric and semi-parametric models, lending support to the effectiveness of the proposed framework.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper provides a strong methodological contribution by combining semi-parametric estimation with multiple realized volatility measures via joint loss functions and adaptive MCMC. The empirical study is highly rigorous, utilizing over 22 years of multi-market data through major crisis periods including COVID-19 to validate the forecasting performance of joint VaR and ES.

Research Flowchart

  flowchart TD
    A[Goal: Enhance VaR & ES Financial Risk Forecasting] --> B[Input Data: 6 Stock Markets 2000-2022 & 3 Realized Measures]
    B --> C[Framework: Semi-Parametric Realized Exponential GARCH]
    C --> D[Likelihood Formulation: Asymmetric Laplace Distribution & Joint Loss]
    D --> E[Model Estimation: Adaptive Markov Chain Monte Carlo MCMC]
    E --> F[Outcomes: Superior 1-Step-Ahead VaR & ES Forecasts vs Benchmarks]

Cited References (33)

  1. Coherent measures of risk (1999)
  2. A Bayesian approach for more reliable tail risk forecasts (2023)
  3. A Statistical Recurrent Stochastic Volatility Model for Stock Markets (2022) — link
  4. Density-tempered marginalized sequential Monte Carlo samplers (2015)
  5. \textup MAR Calculation of \textup RWA for market risk (2023) — link
  6. Autoregressive conditional heteroscedasticity with estimates of the variance of \textup U nited \textup K ingdom inflation (1982)
  7. Generalized autoregressive conditional heteroskedasticity (1986)
  8. Regression shrinkage and selection via the lasso (1996)
and 25 more
  1. Conditional heteroskedasticity in asset returns: A new approach (1991)
  2. On the relation between the expected value and the volatility of the nominal excess return on stocks (1993)
  3. Efficient Metropolis jumping rules (1996)
  4. Answering the skeptics: Yes, standard volatility models do provide accurate forecasts (1998)
  5. Modeling and forecasting realized volatility (2003)
  6. Realized range-based estimation of integrated variance (2007)
  7. Measuring volatility with the realized range (2007)
  8. Realized kernels in practice: Trades and quotes (2009)
  9. Power and bipower variation with stochastic volatility and jumps (2004)
  10. Realized \textup GARCH : a joint model for returns and realized measures of volatility (2012)
  11. Exponential \textup GARCH modelling with realized measures of volatility (2016)
  12. \textup CAV ia\textup R : Conditional autoregressive value at risk by regression quantiles (2004)
  13. Higher order elicitability and \textup O sband’s principle (2016)
  14. Forecasting value at risk and expected shortfall using a semiparametric approach based on the asymmetric \textup L aplace distribution (2019)
  15. Semi-parametric dynamic asymmetric \textup L aplace models for tail risk forecasting, incorporating realized measures (2020)
  16. A semi-parametric conditional autoregressive joint value-at-risk and expected shortfall modeling framework incorporating realized measures (2023)
  17. Risk-parameter estimation in volatility models (2015)
  18. Estimation risk in \textup GARCH VaR and \textup ES estimates (2008)
  19. Goodness of fit and related inference processes for quantile regression (1999)
  20. Dynamic quantile function models (2022)
  21. Weak convergence and optimal scaling of random walk \textup M etropolis algorithms (1997)
  22. Oxford-Man Institute’s realized library, version 0.3 (2009)
  23. The model confidence set (2011)
  24. \textup MFE MATLAB function reference financial econometrics (2009)
  25. Quantile forecasts of financial returns using realized \textup GARCH models (2012)

0 of 33 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.