Natural Disasters And Global Stock Markets

Financial market is considered as a typical complex system consisting of a large number of interacting individuals.With the gradual deepening of global economic integration,the stock market,as the core component of the financial market,will have a huge impact on the world economy in case of a financial crisis.Therefore,it is important to pay attention to the complex network constituted by the global stock market and its systemic risks.On the other hand,due to global warming and climate change,the frequency and intensity of natural disasters such as extreme weather events are increasing,significantly reducing the stability of economic activities.Based on this,firstly,this paper constructs a multilayer dynamic network of global stock markets from the perspective of complex networks by conducting a nonlinear Granger test of order 1 to 12 based on the Tn statistic for the residuals of stock means,variances,and skewness,using stock variables from 20 countries(regions)around the world as the research sample.In-depth analysis is conducted in six aspects:the overall level of network association,the systematic importance of nodes,the association between developed and developing economies,the performance of China in the network,and the spatial spillover effect based on the weighted matrix of nonlinear association.Second,this paper explores the impact of natural disasters on the mean layer,variance layer,and skewness layer of the nonlinear Granger causal network of the global stock market using two dimensions of three natural disaster indicators for risk exporters and inputs.Findings of the study regarding the topology of the multilayer dynamic nonlinear Granger causal network of the global stock markets:(1)the mean network is a polycentric balanced bivariate correlation network;the variance network is a unidirectional output network centered on the dual-core output side;and the skewness network is a loose input network centered on the dual-core nodes.(2)In the short run,the mean nonlinear spillover effect of the stock network is more significant,while in the long run,the variance nonlinear spillover effect is more significant.(3)Korea,Chinese Hong Kong,Australia and Sweden are the major risk exporters;Germany,Spain and France are the major risk importers.(4)The mean of developed economies has more significant risk spillover to developing economies in the short run;developed economies are net risk importers of the three-tier network in the long run.(5)China shows unidirectional risk spillover to the rest of the countries(regions)in the mean and variance layers and is not vulnerable to risk shocks;in the skewness layer,China is vulnerable to risk shocks from nodes such as Japan.(6)The global stock market spatial spillover model based on the nonlinear Granger correlation weighting matrix has significant spatial spillover effects only in the shortand medium-term and long-term.Findings of the study regarding the shocks of natural disasters on multilayer nonlinear Granger causal networks:(1)In the mean layer,natural disasters on the output side predominantly promote the contagiousness of the network;on the input side,they predominantly reduce the contagiousness.(2)In the variance layer,the number of fatalities and the number of affected people on the output and input sides increase the risk contagion of the variance network overall;the economic loss on both sides decreases the degree of association in the opposite direction.(3)In the skewness layer,the number of fatalities and economic losses on the output side reduce the correlation level of the skewness network,and the number of affected people show some boosting effect in the middle and later stages;natural disasters on the input side play a role in continuously reducing the risk contagion of the skewness network.This paper uses the frontier nonlinear Granger test to construct a network that can reveal the complex nonlinear risk propagation path in the financial system;in addition,this paper introduces natural disasters as a new shock variable that can complement the studies related to the impact of natural disasters on capital markets to a certain extent.