| The modern economy often has large fluctuations in asset prices,and this cannot be explained by changes in economic fundamentals.Usually we call these situations or phenomena the expansion and bursting of the bubble.From most of the literature,these bubbles are unpredictable and produce substantial macroeconomic effects.As the bubble expands,consumption,investment,and productivity growth soar;then when the bubble bursts,the economy collapses or stagnate.Therefore,this paper intends to discuss the following questions by studying the bubbles that appear in different asset markets: Can these bubbles be measured? How is the bubble transmitted through space? Is there a linkage effect between the bubbles? This paper mainly uses Log-Period Power Law Model(LPPL)and LPPL based on Quantile Regression to model and measure asset price bubbles,and combines some cutting-edge empirical methods to combine Chinese stock market and China.The real estate market and the characteristics of the international digital currency market,further examine the linkage effect of these asset price bubbles in the quantile measurement,spatial contagion and different markets.In the classic literature of financial market bubbles,the price of financial assets rose sharply in the course of months or even years,far exceeding the reasonable valuation of future cash flows of assets.These price increases were accompanied by a lot of speculation and high trading volume,and the price bubble eventually ended in a crash,in which the price collapse was even faster than the rise.This paper first reviews the research theories and literature of the financial market bubble.Secondly,based on the analysis and review of the existing literature,the LPPL model and the LPPL method based on quantile regression are used to define and segment the stock price bubble and market collapse,and then fit and Research and forecast the financial asset price bubble.The main conclusion is that in the LPPL model,the Confidence index is reliable when fitting the stock market bubble.At the same time,it is found that the empirical analysis shows that the CSI 300 and CSI 500 have larger Confidence values,which indicates that the model fits better in these two markets and can better predict the occurrence of financial asset price bubbles and will be divided.The number regression model is combined with the LPPL model to optimize the prediction effect of the model,and the model is applied to the Chinese stock market to verify the prediction results and enhance the robustness of the model conclusions.At each quantile level,although there is a certain degree of forward or backward review in time,the LPPL model combined with effective data can well predict the outbreak of financial asset price bubbles,and the relevant index can also be somewhat It has an early warning effect on the bubble burst.Again,this paper discusses the price bubble in China's real estate market.The purpose of this part of the research is to establish a real-time price bubble spatial infection model to empirically analyze the spatial contagion of the real estate market in China,and evaluate the local government real estate regulation policies.Specific exploration: How should the price bubble in the real estate market be measured compared to other financial assets? Is there a spatial contagion in the price bubble in the real estate market? What is the impact mechanism behind it?Are the real estate macro-control policies in various places effective to prevent the continued expansion of the real estate price bubble? The research on the above problems can not only quantitatively analyze the real estate price bubbles in various cities in China,but also discover the linkage characteristics between real estate bubbles.The answers to the above questions also help local governments to conduct real estate regulation and control according to local conditions,and provide policy recommendations and references for effectively preventing real estate regional risks.Unlike the stock market,the real estate price bubble is a medium-and long-term price increase process that continues to rise,and the LPPL model can better simulate the process of real estate price growth and reversal(see Zhou and Sornette(2008),Ardila et al.(2013,2017)and Qun,Sornette and Zhang(2017)).Different from the existing literature,this chapter further considers the characteristics of real estate price bubbles in the positive and reverse foam regions.The biggest difference between the two is whether the price dynamics are before or after the price collapse point: the positive bubble is that the price appears faster than the exponential growth and is accompanied by the oscillation,and the price collapse point appears at some point in the future;the reverse bubble appears at the price collapse point.After that,the price is adjusted from bottom to top.By using the real estate market data of 100 cities from June 2010 to November 2017,we conclude that the LPPL model can identify the real estate price bubbles in 100 cities in China and there are mainly two bubble states: positive bubble(House prices continue to rise)and reverse the bubble(the overall decline in house prices has a reversal point).The real estate prices in various cities(regions)have strong spatial contagiousness;the spatial contagiousness of the positive bubble area is more obvious than that of the reverse foam area.Considering the economic space measurement rather than the physical space measurement,the cities The space between them is more contagious.Different from the existing literature,we find that the price index of new homes in the reverse bubble area,especially the rise of the second-hand house price index,has a strong positive impact on the real estate price index of surrounding cities.Finally,we find that the city's real estate regulation and control policy has restrained the influence of the traditional influence of housing prices(such as credit and new housing,second-hand housing prices,etc.)to some extent.However,the linkage characteristics of real estate prices in different cities should be more The attention of the regulator.Finally,this thesis conducts an empirical study on the digital currency that is considered by the industry as “the most bubble”,explores whether there is a bubble accumulation process in digital currency assets,and uses monetary policy as an exogenous shock factor to examine the monetary policy on digital currency asset price bubbles.The linkage effect between.However,the number of existing electronic money assets is more than 400 hours.But the logic and community behind it are basically derived from the three digital currency assets selected in this chapter,and the current market value of these three digital currencies is the top three in the world.On the other hand,although China strictly prohibits the issuance of digital currency,this part of the research still has certain reference significance for regulators.For example,due to the increasing market value and transaction volume of digital money,its influence on other financial markets is gradually increasing.The increase,especially the impact on global financial security,continues to increase,and the ups and downs of digital currency prices will pose a threat to the stability of global financial markets.This section believes that there is a strong correlation between the confidence of the digital currency(the three representative data selected is Bitcoin,Rising Coin and Greeco),which means that there is a gap between the bubbles of different digital currencies.Strong connections,and the price has a limited impact on the confidence of the bubble and the bubble.We have come to the conclusion that the foam index constructed in this part and the price forecasting model of the digital currency asset used can achieve better prediction results by controlling the scope of the training set.At the same time,in the study of the spillover effect of price fluctuations,it is found that the bubble confidence linkage relationship between different digital currencies has its own characteristics.Monetary policy can significantly affect the volatility of the bubble confidence.In the period after the monetary policy is announced,the price of the digital currency tends to fluctuate more violently,and it is more prone to bubble and bubble burst,while the same monetary policy is for different digital currencies.The extent of the impact is different.The main innovations of this paper are:First,by combining the quantile regression model,the traditional LPPL model is improved to optimize the forecasting effect of the asset price bubble.We apply this extended model to the Chinese stock market,verify the prediction results,and find that the model combined with effective data can well predict the outbreak of financial asset price bubbles,and the relevant index can also alert the bubble burst to some extent.effect.The second is to quantitatively measure the real estate price bubble in China,and consider the contagion effect of the real estate price bubble in different spatial(physical and economic)structures.This chapter is based on the basic economic background data of 100 cities at the micro level.Different from the existing literature,this section finds that the new house price index in the reverse bubble area has a strong positive impact on the real estate price index of surrounding cities.The city's real estate regulation and control policy has restrained the traditional influence of housing prices to a certain extent.However,the linkage characteristics of real estate prices in various cities should be more concerned by regulators.The third is to select the most representative digital currencies in the world,use the quantile regression LPPL model to measure the digital currency price bubble,and use the EGARCH model to analyze the reaction effect of different digital currency price bubbles on monetary policy.Different from the existing literature on volatility spillover effects,this chapter explores the linkage(overflow)effect between the probability of occurrence of digital currency bubbles through the confidence indicators derived from the model,and cooperates with monetary policy as an event study to further explore the occurrence of digital currency bubbles.Probability differs between monetary policy before and after. |
PDF Full Text Request