Estimating downside risk in stock returns under structural breaks

Abstract

We show with simulations that inducing structural breaks in the volatility of returns causes non-normality by significantly increasing kurtosis. We endogenously detect significant structural breaks in the volatility of US stock returns and incorporate this information to estimate Value-at-Risk (VaR) to measure the downside risk. Out-of-sample performance results indicate that our proposed model, which incorporates both time varying volatility and structural breaks in volatility, produces more accurate VaR forecasts than several benchmark methods. We highlight the economic importance of our results by calculating the daily capital charges using the Basel Accords.