Taking advantage of proxy variable of stock price manipulation like ownership concentration and earning management to examine the price difference between two market. In addition to liquidity or demand elasticity,etc, using manipulate variable of premium market(A-shares compared to H-share,TDR compared to HK-share) to investigate if manipulation can explain stock price difference. Empricial results show that premium not only concern to market system, but also connect to manipulation factors.The degree of premium will decrease as time progressd.

The price difference of the Chinese dual-listed companies is a interesting issue. The price of the same asset should be consistent after risk-adjusted, but there is an obvious difference between A-share and H-share which the same company¡¦s shares listed in the Shanghai or Shenzhen Stock Exchange and Hong Kong Stock Exchange. This study found that not only liquidity, demand, required risk premium, and information asymmetry can lead to the price difference, but also manipulation, investor¡¦s preference, and the market emotion. This also verified the investors are more speculative in China than the investors in Hong Kong. Besides, this research analysis the price difference ratio form the viewpoint of valuation, the difference ratio provides a way to understand whether the stock is undervaluation or overvaluation. Finally, this study discussed the Taiwan Depository Receipts. The result indicated that most of the factors which influence the price difference between A-share and H-share also affected the price difference between TDR and the original share. The difference ratio between TDR and original share can also provide some information about the stocks are undervaluation or overvaluation.

This thesis gives a brief account on the segmented Chinese stock markets. The indexes of A shares, i.e. shares on the domestic market sold only to domestic investors and the H share index, i.e. foreign shares sold on a foreign stock market, Hong Kong, and sold only to foreigners are compared. With time, the Chinese government is opening up its stock market more and more and allowing more interaction between local and domestic stock markets, but one step at a time. Three major attempts to open up the markets are described and investigated on their effect on the integration of the H share market and the A share market. We checked for the introduction of CEPA, QDII and “through train”. The tests applied were ADF test, Engle-Granger cointegration test and Granger causality test. We found no cointegration for the entire sample and in none of the subsamples we used. The findings on the causality relations among the different stock markets don’t confirm any of the four causality relations defined before. We don’t see any unidirectional causality and it changes over time. We cannot confirm the global center hypothesis or the home bias hypothesis, but we can also not claim that the markets are completely segmented and show no correlation among prices. There is a correlation and there are causality links but they change every time the government changes its regulations. However, it is not clear what kind of regulations will make the causality change in which direction.

China's rapid economic growth and the development of its domestic stock market have attracted considerable attention from foreign investors. China's economic financial expansion, however, has emerged from an environment of state planning and radical socialist ideology. With a view of providing investors with a better understanding of the risk and return relationship in the Chinese A-share market over the past decade, this thesis adapts several empirical models to the circumstances in China and conducts four empirical analyses. First, in order to rationalize foreign investors' entry into the A-share market, the thesis compares the diversification benefits in three China-related stock markets, namely the A-share, the B-share and the H-share markets in a mean-variance framework using daily, weekly and monthly data respectively. The results suggest that of the three stock markets, the B-share market generates the highest average annual returns while the A-share market has the most significant diversification benefits regardless of whether the analysis is undertaken implementing a traditional mean-variance framework or a downside risk framework. Next, an empirical analysis using the Fama and MacBeth two-pass procedure is undertaken to test the relationship between beta, firm factors and stock returns. Similar to the findings in other stock markets, the results of this analysis show that the static betas for individual stocks fail to capture variation in stock returns in the A-share market. In contrast, the effects of book-to-market and trading volume are significant in the sample period. However, the fact that none of these factors have a persistent role in explaining stock returns suggests a possible change in the investment philosophy of Chinese domestic investors over the past decade. In the third analysis, two global betas are incorporated into the cross-sectional regressions in a bid to examine the integration or segmentation of the A-share market with the world and Hong Kong stock markets. Specifically, both time-varying betas and static betas are used in the analysis. The results suggest that there is no beta effect and the A-share marke t is totally segmented from both the world and Hong Kong stock markets. Finally, when the segmentation and integration status of the A-share market is further examined using the Maximum Likelihood Estimation framework without beta estimation and the assumption of a linear relationship between beta and stock returns, the findings suggest that the A-share market is becoming increasing integrated with the B-share and the Hong Kong stock markets.

China's rapid economic growth and the development of its domestic stock market have attracted considerable attention from foreign investors. China's economic financial expansion, however, has emerged from an environment of state planning and radical socialist ideology. With a view of providing investors with a better understanding of the risk and return relationship in the Chinese A-share market over the past decade, this thesis adapts several empirical models to the circumstances in China and conducts four empirical analyses. First, in order to rationalize foreign investors' entry into the A-share market, the thesis compares the diversification benefits in three China-related stock markets, namely the A-share, the B-share and the H-share markets in a mean-variance framework using daily, weekly and monthly data respectively. The results suggest that of the three stock markets, the B-share market generates the highest average annual returns while the A-share market has the most significant diversification benefits regardless of whether the analysis is undertaken implementing a traditional mean-variance framework or a downside risk framework. Next, an empirical analysis using the Fama and MacBeth two-pass procedure is undertaken to test the relationship between beta, firm factors and stock returns. Similar to the findings in other stock markets, the results of this analysis show that the static betas for individual stocks fail to capture variation in stock returns in the A-share market. In contrast, the effects of book-to-market and trading volume are significant in the sample period. However, the fact that none of these factors have a persistent role in explaining stock returns suggests a possible change in the investment philosophy of Chinese domestic investors over the past decade. In the third analysis, two global betas are incorporated into the cross-sectional regressions in a bid to examine the integration or segmentation of the A-share market with the world and Hong Kong stock markets. Specifically, both time-varying betas and static betas are used in the analysis. The results suggest that there is no beta effect and the A-share marke t is totally segmented from both the world and Hong Kong stock markets. Finally, when the segmentation and integration status of the A-share market is further examined using the Maximum Likelihood Estimation framework without beta estimation and the assumption of a linear relationship between beta and stock returns, the findings suggest that the A-share market is becoming increasing integrated with the B-share and the Hong Kong stock markets.

We develop a non-linear Markov error correction approach to examine the general co-integration relation between the H- and A-prices of cross-listed Chinese stock issuers across the period January 1999 to March 2009. We unravel three important dimensions of this relation. These pertain to (i) the long-run expectation of the H- (to A-price) discount; (ii) the level of short-run co-movement in prices; and (iii) the magnitude of error corrections. Findings point to significant improvements in all three areas. Policy and corporate governance change appears to be the principal force driving the efficiency gains. Weakening informational asymmetries underlie much of the change in the markets’ relative pricing. In contrast, sentiment effects strongly underpin the contemporaneous response and error correction adjustments. Finally, the escalating Global Financial Crisis of 2008 appears to have not only bolstered the A- and H-markets’ short-term pricing dynamic but also temporarily increased the long-term H-share discount.

Cette thèse se compose de trois essais autonomes sur le marché boursier chinois. Le premier essai examine le processus de la découverte du prix des actions A et H pour des sociétés chinoises double cotées à la fois sur les bourses de Shanghai/Shenzhen et de Hong Kong durant les sessions d'échange communes. Nous mettons en évidence une relation de long terme entre les prix des actions A et H. En appliquant la méthode de l'information partagée de Hasbrouck (1995), il apparaît, quand la Chine adoptait un régime de change fixe, le marché domestique contribuait plus d'information à la découverte du prix que le marché étranger; tandis que sous un régime de change flexible, c'est le marché étranger qui dominait dans la découverte du prix.Le deuxième essai prenant les réformes chinoises du régime de Juillet 2005 et de Juillet 2008 comme des événements spéciaux, il étudie si ces changements de régime de change affectent l'arbitrage entre les marchés des actions A et H. En comparant les niveaux des impacts des facteurs idiosyncratiques sur la décote de prix des actions A et H avant et après les changements de régime, les résultats montrent que la relaxation des contrôles des changes ne favorise pas l'arbitrage entre les deux marchés. Par ailleurs, ce changement de régime de change introduit un risque de change important dans la stratégie des arbitragistes.Le troisième essai aborde la transmission d'information en séance et hors séance de cotation en termes de rendements et de volatilités entre la Chine, l'Amérique et l'Europe. Le problème du synchronisme est considéré avec soin dans la modélisation bivariée avec la Chine comme référence avec des données journalières. / This thesis consists of three self-contained essays on the Chinese stock market. The first essay examines the price discovery process of Chinese dual-listed firms on the A-share and H-share markets during overlapping trading hours. We provide evidence that there exists a long-term relationship between the A- and H-share markets. By applying the information share model of Hasbrouck (1995), we find that: under a fixed exchange rate, the A-share market contributes more innovations in price discovery than the H-share market; while under a managed floating exchange rate, it is the H-share market that plays a dominant role in the price discovery process.In the second essay, by using the exchange rate regime changes of July 21, 2005 and July 01, 2008 of as special events, we examine whether changes in exchange-rate regime affect the intensity of inter-market arbitrage between A- and H-share markets. By comparing the significance of the impact of idiosyncratic factors on the H-share discount before and after the changes of exchange rate regime, the results show that the relaxation of exchange controls does not encourage inter-market arbitrage between the Chinese mainland and Hong Kong markets. Further, the switch from a fixed to a floating exchange-rate regime introduces an important exchange rate risk to arbitrageurs.The last essay studies daytime and overnight information transmission in terms of returns and volatility between China, America and Europe. The asynchronicity issue is carefully considered in the bivariate modelling with China as benchmark with daily data.