Forecasting risk with Markov-switching GARCH models:A large-scale performance study

2018/08/10 by David Ardia, Keven Bluteau, Kris Boudt +1 · 1 citation
Economics, Econometrics and Finance · #Financial Risk and Volatility Modeling #Market Dynamics and Volatility #Monetary Policy and Economic Impact

paper · pdf · doi:10.1016/j.ijforecast.2018.05.004

openalex publication_date 2018/08/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/02

Abstract

We perform a large-scale empirical study in order to compare the forecasting performances of single-regime and Markov-switching GARCH (MSGARCH) models from a risk management perspective. We find that MSGARCH models yield more accurate Value-at-Risk, expected shortfall, and left-tail distribution forecasts than their single-regime counterparts for daily, weekly, and ten-day equity log-returns. Also, our results indicate that accounting for parameter uncertainty improves the left-tail predictions, independently of the inclusion of the Markov-switching mechanism.

Citations

• Maximum likelihood estimation of the Markov-switching GARCH model
• Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation
• Conditional Heteroskedasticity in Asset Returns: A New Approach
• Generalized autoregressive conditional heteroskedasticity
• Robust forecasting of dynamic conditional correlation GARCH models
• GENERALIZED AUTOREGRESSIVE SCORE MODELS WITH APPLICATIONS
• A Markov Model of Switching-Regime ARCH
• CAViaR
• Autoregressive conditional heteroskedasticity and changes in regime
• Combining density forecasts using focused scoring rules

Cited by

• The M4 Competition: 100,000 time series and 61 forecasting methods