On the predictive power of network statistics for financial risk indicators

Jianhua Song,Zhepei Zhang,Mike K.P. So
DOI: https://doi.org/10.1016/j.intfin.2021.101420
2021-11-01
Journal of International Financial Markets, Institutions and Money
Abstract:An understanding of the financial instability during financial crises is an important topic in risk management. Market participants actively use risk indicators, such as the VIX in the US, the VHSI in Hong Kong and the V2TX in Europe, which are derived from derivative products, to measure market anxiety and fear and thus to estimate systemic risk in the market. In this paper, we present the findings of a study on the lead–lag relationship between financial connectedness and risk indicators. Specifically, we examine the predictive power of time-varying network statistics, compiled from more than 1300 stocks from international stock markets, on the risk indicators. Our empirical findings show strong evidence in favor of using network statistics to predict risk indicators. The findings reveal the importance of network connectedness in measuring systemic risk.
What problem does this paper attempt to address?