Research On The Contagion And Measurement Of Financial Market Tail Risk Based On Macroeconomic Factors

Extreme events have the distinctive characteristic that they have a large or extreme impact when they occur.In the economic and financial fields,extreme events are occurring more frequently than ever before,each generating significant financial market tail risks with serious consequences.Moreover,with the continuous promotion of economic integration,financial liberalization and transaction expedition on a global scale,and the continuous development and updating of new financial interactions such as the Internet economy and electronic quick payment,global financial markets have become more closely connected,and the trend of simultaneous rise and fall among different financial markets has begun to emerge.Hence,once extreme events occur,the resulting financial market tail risks will inevitably produce a domino effect that affects the entire financial system.At present,academics and practitioners have done research on financial market tail risk contagion and measurement,but the existing studies have not included macro and micro information into the same framework for analysis due to the information frequency problem.However,since the 21st century,economies have become increasingly connected and dependent on each other,and the profound impact of macroeconomic fundamentals of a region or country on the global economic environment and financial market stability can no longer be ignored.Part of the literature explores the impact between macro variables and financial variables by unifying their frequencies.The models involved in these literatures require the data to be at the same frequency,which can cause problems such as data distortion and also cannot make maximum use of the valid information contained in data of different frequencies.In this context,there is an urgent need to incorporate macroeconomic factors and financial market variables simultaneously into the consideration of tail risk in financial markets.To fully understand the impact of the macroeconomic conditions on the tail risk spillover and contagion in financial markets,this dissertation discusses the risk spillover and contagion of a country’s macroeconomic conditions on the stock and futures markets based on the mixed data sampling(MIDAS)method,the tail index regression(TIR)model,and the time-varying mixture(TVM)model.Moreover,this dissertation provides a dynamic understanding of the changes in tail risk linkages between different markets during extreme events.In addition,we capture the performance of macroeconomic variables and financial market variables in the tail risk contagion of financial markets and measure the tail risk profile faced by different financial markets.Finally,we propose tail risk avoidance measures for financial markets in order to clarify the role of different macroeconomic factors in the tail risk contagion in the market and promote the sound,smooth and efficient development of financial markets through these results.The specific findings of the study are as follows.First,on the basis of not converting data frequencies,this dissertation can incorporate different frequency data into the same framework by constructing TIRMIDAS models and TVM-MIDAS models,which reasonably portray the effects of macroeconomic factors on the tail behavior of financial markets.In addition,these models can effectively test the financial market risk contagion and accurately measure the corresponding financial market risk.Second,using the TIR-MIDAS model,it is found that there is heterogeneity in the impact of China’s macroeconomic conditions on different BRICS countries.The Brazilian stock market is affected by all macroeconomic variables in China,while South Africa and Russia are vulnerable to China’s PPI and IP.In contrast,the Indian stock market is completely opposite and is not affected by Chinese macroeconomic factors.The results of the parameter estimation also reveal that the Chinese stock market is less influential than the Chinese macroeconomic factors on the stock markets of the BRICS countries.Moreover,the current return changes in the Chinese stock market have no significant impact on all BRICS stock markets.Third,based on risk contagion analysis,the dissertation finds that there are significant differences in the effects of macroeconomic factors and stock indices in the Chinese market for different BRICS stock markets.The Chinese stock market and macroeconomic conditions have risk contagion to the Brazilian and South African stock markets,while there is almost no risk contagion effect to the Russian and Indian stock markets.Fourth,applying the TVM-MIDAS model,we find that the U.S.CPI is insignificant for all tested markets,which implies that it may have no effect on return volatility in all tested markets,while the U.S.IP plays a significant role in all tested economies’ stock markets.Specifically,IP negatively affects Australia,Hong Kong and UK stock market risk,while it positively affects German and Japanese stock market risk.In addition,when the U.S.stock market is booming,stock market risk in the tested economies is reduced.And when the U.S.stock market is depressed,the corresponding stock market risk increases.Fifth,risk contagion from the U.S.market to other economies occurs whenever there is a world-class extreme event.However,there are differences in the strength of risk contagion in the U.S.market.The contagion effect was more pronounced in the U.S.during 2001 and 2008 compared to other periods of extreme events.Sixth,based on the TVM-MIDAS model,we investigate the contagion and measurement of macroeconomic factors on futures market risk in China.The results find that CPI has a significant positive effect on the price volatility of all tested commodity futures,while IP has a catalytic effect on the volatility of corn and coke commodity futures prices.In addition,China’s macroeconomic factors have risk contagion effects on commodity futures such as copper,aluminum,palm oil and coke,while there is no risk contagion effect on corn and fuel oil.Moreover,the TVM-MIDAS model with macroeconomic information improves the validity and accuracy of VaR estimation for commodity futures.Finally,by constructing a VaR-based portfolio,we also find that the TVM-MIDAS model has better performance.