The Research On The Risk Spillover Effect Of The Commodity Market To Chinese Stock Market

With the globalization of world economy,the capital market openness,and governments around the world starting to reduce the intervention to financial markets,the boundaries among the financial markets in different countries and other markets in the same country are becoming more and more blurred.This trend will undoubtedly promote the circulation of capital and optimize allocation of resources,but it also increases the possibility of the risk contagion among different countries and markets.As the biggest emerging market,China has the second largest stock market in the world,and at the same time,China is also the world biggest manufacturing and consuming country of commodities.But it has to be admitted that China's stock market is still at low levels of development and quite vulnerable to risks.As for commodities,China depends on importing heavily.At the beginning of 2018,America restarted the trade protection policy,which increased the potential risk of commodity market.Exactly based on this background,this thesis analyses the risk spillover between China's stock market and the world commodity markets using both theorical and empirical methods.It is expected that the research results might provide some useful guidance for investors and regulators to better understanding the risk spillover between stock and commodity markets.In this thesis,the daily closing price from 2006 to 2018 of Shanghai composite price index and some relative price indexes of commodities are selected as the study sample and the GARCH-Copula-CoVaR is implemented to analyze the risk spillover effect between China's stock market and the world commodity market.Firstly,the common ARCH effect in financial time series is handled by using GARCH(1,1)model and according to the maximum log likelihood and minimum AIC information criteria rules,the skew student t distribution is selected as the optimal marginal distribution.Then,the unconditional VaR of each market is calculated based on the GARCH model results.After that,the risk indexes constructed based on VaR are used to test the risk spillover effect with logit regression method.Secondly,in order to fit the Copula function,the standard residuals of GARCH model are extracted and applied to probability integral transformation.And The optimal Copula functions are also selected according to the log likelihood and AIC rules.Finally,the CoVaR,?CoVaR and %CoVaR are calculated by solving the upper limit of double integral,and then applied to horizontal and vertical comparison.The research results of this thesis suggest that: Firstly,oil and copper both have positive risk spillover to stock market and the spillover intensity of oil is weaker than that of copper.secondly,the spillover test shows no evidence of spillover effect between gold and stock market and thus,the CovaR calculated for gold is the smallest of all.Thirdly,commodity market as a whole has the strongest spillover intensity.And last,the risk spillover intensity between China's stock market and commodity market variates over time.In crisis period,the spillover intensity is obviously stronger.Choosing the commodity and stock market as the research target,this thesis mainly explores the risk spillover between these two markets,and during the empirical analysis the GARCH,Copula and CoVaR model are comprehensively applied to describe not only the volatility of financial timeseries but also the dependent status,and the rick spillover between the two markets is measured accordingly.At the same time,the time-varying characteristic and overall dependent structure of financial timeseries are both taken into consideration,and this is one of the innovations of this thesis.But,when using Copula to measure the dependent status between commodity and stock market,this thesis fails to consider the time-varying characteristics of dependence,and this flaw will be improved in the future researches.