Network diffusion of international oil volatility risk in China's stock market: Quantile interconnectedness modelling and shock decomposition analysis

Abstract

This paper aims to investigate whether oil volatility risks could diffuse through the linkage of industry returns and even further contribute to global crisis throughout the stock market. To highlight such sectoral financial interconnectedness which may serve as a key channel for oil volatility diffusion, we first apply the partial cross-quantilogram (PCQ) approach to detect the directional predictability between returns of 26 industries in China's stock market across different quantiles. We construct the corresponding network to provide a more comprehensive picture of such sectoral interconnection, which is shown to vary prominently under different market states and lag order specifications. Utilizing the spatial autoregressive (SAR) model for panel data, we further assess the possibility of the oil volatility risk diffusion by decomposing the aggregate effect of oil volatility shocks into direct shocks, and indirect shocks transmitting through the estimated network linkage of industries. The empirical results point to the significantly heterogeneous pattern of oil volatility risk diffusion among networks with different lag selections under various market states. Considering the financial linkage of industries within 5 traded days under extreme market states, there exist significant indirect effects contributing to larger oil volatility shocks on industry returns, which confirms risk contagion effects of monthly oil volatilities. It indicates that the network linkage of financial assets might be an important diffusion mechanism for oil volatility risks, which should not be neglected in the research of the oil-stock nexus.