Comovement Of Chinese Stock Market

Comovement is a common phenomenon in security market, which refers to the assets prices move in the same direction. This comovement can occur in both between the global stock markets and in one market. Since the 1990s, comovement effect gradually strengths between different countries and regions, especially before and after the 2008 financial crisis, many countries'stock markets have experienced from the Skyrocketing growth to the overall decline. In addition, there are relative effects of return comovement among industry parts in one country. For example, the aviation of 1950s and the network section of 1990s. Therefore, we call the same direction among different markets, regions, industries and plates as comovement effect.There exist two different views of return comovement:the traditional view, which attributes it to comovement in news about fundamental value, and the alternative view- the theory of behavioral finance, in which frictions or sentiments delink it from fundamentals. The theory of behavioral finance broadens the "rationality assumption" and points out the stock market is not efficient.On the basis of the theoretical hypothesis and researches at home and abroad, our paper will explore the comovement of China's stock market. We study CSI 300 and CSI 100 to provide evidence in support of the friction-or sentiment-based theories of comovement. CSI 300 Index is composed of 300 largest and most liquid stocks selected from the Shanghai and Shenzhen stock exchange. CSI 100 is composed of 100 largest stocks in CSI 300. These two good running indices have been accepted by investors that supply sufficient data for study. We can study them to provide evidence in support of Chinese stock market existing comovement.We divide the paper into three parts:Part one contains the former three chapters which are introduction, literature review and theory & model, respectively. Chapter 3 firstly introduces relative theories which are consisted of Asymmetric Information, Efficient Market Hypothesis and Behavioral Finance. Secondly, we briefly present reduced-form models of three friction- or sentiment-based theories of comovement: the category, habitat, and Information diffusion views and give the mathematical formulas of these three views. Finally, establish the univariate regression and bivariate regression.Part two contains chapter 4 and chapter 5 which is the main section. Chapter 4 mainly introduces the characteristic, computation and adjustment program of the CSI 300 and CSI 100. It introduces both the sample selection and the methods of data processing as well. In chapter 5, we conduct an empirical analysis on the comvement of CSI 300 and CSI 100.Part three is the conclusion and future work.We present two broad theories of return comovement and examine them empirically using data on additions and deletions to the CSI 300 and CSI 100. The paper makes three findings in the following aspects.First, we find that stock added to the CSI 300 or CSI 100 experiences a significant increase in beta after inclusion. We then show that in bivariate regressions that also include the return of non- CSI 300 or non- CSI 100, the rise in CSI 300 or CSI 100 beta is altogether larger than anything in the univariate regressions, and that exactly the opposite pattern holds for stocks deleted from the CSI 300 or CSI 100. We also find that in both univariate and bivariate regressions, the effects of deletions are stronger than that of inclusions. Lastly, we test the return comovement between the CSI 300 and the CSI 100, the result is that there have Long-term and significant comovement among CSI 300 and CSI 100. In addition, the Granger causality test shows that the CSI 300 is the Granger cause of the CSI 100. Our findings cannot easily be explained by the fundamentals-based view and provide new evidence in support of the friction- or sentiment-based view.This paper makes works are as follows. First, in the sample selection, CSI 300 have 256 inclusion events and 361 deletion events in the sample period; CSI 100 have 54 inclusion events and 81 deletion events in the sample period. Second, inclusion (deletion) events are excluded if the new firm is a spin-off or a restructured version of a firm already in the index, if the firm is engaged in a merger or takeover around the inclusion (deletion) event, or if the event occurs so close to the end of the sample that the data are not available. Third, in the respect of time window, we exclude the stock that is in the month of the event and that is less than 200 trading days. Forth, it divided the period into some subset periods according the 2008 financial crisis. Finally, we add the CSI 200 to examine the comovement of CSI 100 and CSI 300.The shortcomings still exist in the paper. There are too less studies about comovement at home so that the literature reviews are not perfect. Next, we only apply the calendar time test causes the paper is not very persuasive. We failed to provide direct empirical support on between the noise trader sentiment and comovement. Future research will work on solving these problem above.