Asymmetric volatility in equity markets around the world

The observation that price declines usually lead to volatility increases is known as the asymmetric volatility effect and has become a stylized fact about the financial markets. We study asymmetric volatility effect in 19 equity indices from North America, Latin America, Europe, Asia and Australia, utilizing not only daily data and four GARCH class models, but also realized volatility calculated from high-frequency data within HAR class models. We first confirm the stylized fact that stock market indices around the world exhibit the asymmetric volatility effect. This effect is stronger for US and European market indices. Second, we find that the asymmetric volatility effect is strong enough to significantly improve out-of-sample forecasts of an accurate HAR volatility model. Third, we show that forecast improvements of the asymmetric volatility models are largest during periods of higher market volatility, when accurate volatility forecasts matter the most.