The meaning of structural breaks for risk management: new evidence, mechanisms, and innovative views for the post-COVID-19 era

Abstract

This paper quantitatively reveals the meaning of structural breaks for risk management by analyzing US and major European banking sector stocks. Applying newly extended Glosten-Jagannathan-Runkle generalized autoregressive conditional heteroscedasticity models, we supply the following new evidence. First, we find that incorporating structural breaks is always effective in estimating banking stock volatilities. Second, we clarify that structural breaks partially explain the tail fatness of banking stock returns. Third, we find that when incorporating structural breaks, the estimated volatilities more accurately capture their downside risk, proving that structural breaks matter for risk management. Fourth, our news impact curve and model parameter analyses also uncover that when incorporating structural breaks, the asymmetry in volatility responses to return shocks is more accurately captured. This proves why the estimated volatilities by incorporating structural breaks better explain downside risk. In addition, we further reveal that the estimated volatilities obtained through incorporating structural breaks increase sharply during momentous events such as the Lehman crisis, the European debt crisis, Brexit, and the recent COVID-19 crisis. Moreover, we also clarify that the volatility spreads between models with and without structural breaks rise during the Lehman and COVID-19 crises. Finally, based on our findings, we derive many significant and beneficial interpretations, implications, and innovative views for risk management using artificial intelligence in the post-COVID-19 era.