Dynamic risk spillover among crude oil, economic policy uncertainty and Chinese financial sectors

Abstract

Using daily return, realized volatility, realized skewness and realized kurtosis as risk proxies, we analyze the risk spillovers among crude oil, gold, economic policy uncertainty (EPU) and four Chinese financial sectors including bank, trust, insurance and security under different market conditions. Our analysis spans from January 10, 2008 to June 30, 2020 and is based on the combination of quantile VAR model and time-varying parameter vector autoregressive (TVP-VAR) model based on generalized forecast error variance decomposition. We find that: (1) The risk spillovers of the four risk proxies under 0.01-quantile and 0.99-quantile are much larger than those under mean and median (0.5-quantile); (2) The four risk proxies capture different information of the systemic shocks; (3) Term spread and credit spread have significant forecasting power on the total return spillovers and the total volatility spillovers.