Research On The Complex Mechanisms Of Uncertainty And Asset Price Dynamics

Asset pricing is one of the classical problems in finance,and its status in finance is like the general equilibrium in economics,which means that it attracts millions of academics and practitioners.The financial logics hide from series of financial and investment researches,and how to solve the puzzle of asset pricing has drawn much attentions.From the Mean-Variance Model by Markowitz?1952?,to Capital Asset Pricing Model,and Efficient Market Hopothesis,Black-Scholes formular,Behavioral finance and Informative finance,the financial asset pricing model experienced a quick development with a scientific research structure.With many asset pricing factors revealed,the key of asset pricing has changed,and the tests of traditional factors and research on the new figures of capital markets become the mainstream.However,revealing new factors is inevitably the pursuit of the financial researchers.The financial crisis caused heavy damages to the world economy and financial markets.The developed countries firstly started irregular monetary policies after the crisis,in order to help the economy,stabilized the markets,and save the employments.The developing markets kept a strong economic growth.However,as the financial crisis went on,the implied uncertainty was revealed,all the countries were stuck by the crisis.The question that where to go is not answered.In the complex world background,the Chinese economy is now in a period of growth changes,power changes,and structural adjustments,and experiencing magnificent difficulties and challenges,like excess manufacturing capacity,excess inventory,debt default,supply-demand mismatch.More importantly,the financial institutions being distracted from the real economy is now a classical problem.The trade war is an important factor that triggers the uncertainty of the future economy.The only path for this problem is the reform and open policy,which is also difficult in this situation.As the world's economy and finance go to an unexpected direction,there are many challenges on the financial asset pricing.Can the traditional asset pricing factors found in stable economies be adaptive in the nowadays economic environment?This is a question that is controversial.Many people started the research on the impacts of political disagreement,natural disasters and area conflicts on the asset returns and volatility.These factors all can be seen as uncertainty.Considering the world economy and the research situations of financial asset pricing,it is not difficult to understand the relationships between uncertainty and asset pricing become the key point.The economic uncertainty contains more pricing information,including macroeconomic and financial fundamentals,economic policies,and enterprise functions,it also contains the expectations of the economists and investors for the future.The complex and overall means of uncertainty have made it a powerful asset pricing factor.Based on the summarization of the previous related literature,we can find that the research structure of the relationship research between uncertainty and asset pricing is not yet been done,and there is no enough theoretical fundamentals.Considering the relationship between uncertainty and asset pricing is a new research field,we find that the researches are not abundant,and some new areas need to be explored.So,this thesis tries to research on the dynamic relationships between them from several aspects.We use main stock market data,and systematically explore the relationships between uncertainty and asset pricing from the perspectives of predictability,spillover,and nonlinearity,based on many econometric models.The research details and structures in this thesis are presented as follows:The 1^st chapter is Introduction.This chapter presents the research background and meanings,basic definitions,main research details,innovations and shortcomings.The 2^nd chapter is Literature review.This chapter summarizes the related literature and makes comments on them,for the start of the research of this thesis.We summarize the related literature from these three aspects:1.Asset pricing.We summarize the factors that could influence asset returns and volatility.2.Uncertainty.We find the research fields and structures of uncertainty research.3.Combinations both of them.We summarize the research from the theoretical and empirical aspects.The 3^rd chapter is Theoretical framework.This chapter shows the theoretical fundamentals of the impact of uncertainty on asset returns and volatility,and tries to extend the existed framework.We first distinguish the uncertainty indices into two categories:general economic uncertainty and limited attention uncertainty,based on the principals and basic data.We then discuss the impact mechanisms of uncertainty on returns and volatility.We also try to give a channel of the impact of uncertainty on volatility based on the countercyclical pattern of financial volatility:uncertainty could affect the economic activity,and higher uncertainty would lead to higher volatility according to the countercyclical pattern.The 4^th chapter is the Empirical Analysis of the predictability of uncertainty.Considering the lack of limited attention uncertainty,this thesis builds a new limited attention uncertainty based on search volume index and keywords related to uncertainty.Based on the keywords,we build three kinds of uncertainty:policy,stock and macro.We then explore the predictability of the new uncertainty on the main markets,using Granger Causality test,linear regression and out-of-sample analysis.The 5^th chapter is the Empirical analysis of uncertainty spillover.This chapter tries to analyze the relationships from the aspect of spillover,which helps to provide a new perspective for market comovement and risk contagion,and also provide new evidence for the relationships between uncertainty and asset price volatility.This chapter researches on the spillover of US uncertainty on the other main market volatility with the two-factor GARCH-MIDAS model extended from single-factor GARCH-MIDAS model.The 6^th chapter is the Spillover of uncertainty for China.This chapter tries to find out the uncertainty spillovers of the main countries to the Chinese stock market,in order to reveal the most important uncertainty sources,and help to the policy decisions for the Chinese government.We build a new model that combines variable selection with GARCH-MIDAS model,and choose the most important uncertainty index that could impact the Chinese stock market volatility.In this aspect,this chapter also contributes to the econometric theory.In addition,we also explore the spillover of Chinese uncertainty on the other markets using two-factor GARCH-MIDAS model.This chapter has innovation of econometric models and research aspect,which is a key chapter of the thesis.The 7^th chapter discusses the nonlinear relationships between uncertainty asset prices.This chapter,from the perspective of nonlinearity,considers 16 stock markets,including the developed and developing markets from America,Europe,Asia and Pacific,to present a relatively systematic research.This chapter uses nonlinear Granger causality to test the nonlinear relationships.The empirical analysis contains the nonlinear relationships between market returns and uncertainty,and market volatility and uncertainty.We also discuss the structural characters based on structural breakpoints tests with subsample analysis.The 8^th chapter concludes,and proposes the policy suggestions and the future possible research.The main empirical conclusions are as follows:1.The impacts of limited attention uncertainty on the world main markets are different across different areas.The impact of such uncertainty on the returns is asymmetric.In the out-of-sample evaluations,the performances are different for the different uncertainty,and it also has area differences.2.The US EPU spillovers to the developed markets are positive,and to the developing markets are negative.The NVIX spillovers to the developed market volatility are more significant after the financial crisis.3.The most important indices that influence the Chinese stock market are German EPU and French EPU,and the impact of North American EPU is overestimated.The Chinese own EPU still has the biggest impacts.The Chinese EPU has negative impacts on the other markets,and the impact is significant for France,Japan,Russia and Brazil.4.For most markets,there is no nonlinear relationship between EPU and stock market returns,which indicates the relationships are more in the aspect of linearity.The stock market volatility is more sensible to EPU,and EPU can be seen as an endogenous factor.The relationships between EPU and markets are different across the developed and developing markets.Based on the main conclusions above,this thesis proposes these suggestions:we should determine the most important uncertainty based on economic connections;we should build the real Chinese economic uncertainty based on big data;we can see that it is not necessary to build very complex systems for uncertainty alert;we should forward the capital market construction,in order to increase the ability of defend the outside risks.This thesis thinks that research can be considered in these areas in the future:the theoretical model can be further improved,more new methods for uncertainty indices are explored,and the endogeneity of uncertainty can be tested.