The Research On Spillover Effect Between A Shares And H Shares

A+H cross-listed stocks have the same subject. Therefore, there is a natural link between prices of them. With Shanghai-Hong Kong Stock Connect program being launched, the correlation between prices of A shares and H shares is being increased. However, due to the two different trading mechanisms and many other factors, there is still a big difference in prices of them. Besides, yield of A shares often deviate from the corresponding H shares. Without price limit, H shares can reflect the true value of the company faster in the event of a burst of good or bad news while A shares may hit the price limit continuously in the following days. Study on the above spillover effects will help investors to make investment decisions and regulatory authorities to set policies.We mainly use event study method in this thesis. By calculating the correlation coefficients, back-testing historical data, using comparative analysis and Granger causality test, we conduct empirical research and draw the conclusion that A shares demonstrate spillover effect after prices of corresponding H shares rise by more than 15%. The main contents are as follows.The first two chapters are the introduction and literature review.In chapter three, we compare A share market with H share market and introduce the evolution about price limit of mainland stock market. We compute Pearson correlation coefficients between A and H shares based on the historical data. Meanwhile, we compute Pearson correlation coefficient among 2105 stocks in the same period. By comparing the two groups, we prove that A and H shares have significant positive correlation with sound statistics.In chapter four, we first back-test average cumulative returns of corresponding A shares in the following sixty days after prices of H shares rise by more than 15% based on the data from January 1st in 1997 to August 19th in 2015. By comparing the result with that of all the A shares in the same period, we find that corresponding A shares demonstrate spillover effect. What’s more, the effect will be more significant if corresponding A shares hit the limit price. Second, we select 2889 historical examples with similar price change of Shanghai composite index and that of A shares to those when prices of corresponding H shares rise by more than 15%, and then back-test average cumulative returns in the following sixty days. The result shows that Shanghai composite index and A share’s own price change don’t result in spillover effect of A shares. This thesis’s further study shows that in the following sixty days after prices of corresponding H shares rise by more than 15%, A shares obviously perform better than the control group similar to trends of the former in the past sixty days. Using the same method, we back-test average returns of corresponding A shares and all the A shares after prices of H shares fall by 15%. The result shows that the event that prices of H shares fall by 15% has negative impact on prices of corresponding A shares while the impact is less significant than that of the event that prices of H shares rise by 15%.In chapter five, We back-test cumulative returns of H shares after prices of H shares rise by more than 15%. We compare the result with that of all the H shares and find there is no significant difference between the two. Finally we draw the conclusion that it is H share’s big price changes that lead to spillover effect of A shares. The results of Granger causality test show that if H shares have stronger effect on corresponding A shares, returns of these A shares will perform better in the following days.In chapter six, from the perspective of the deal, we find that there is an 80% chance that the return will be 5% and a 70% chance that the return will be 10% if we buy the corresponding A shares after prices of H shares rise by more than 15%. Finally, we analyse some examples of buying shares according to our finding and give some useful suggestions.Investors who make investment decisions by using the above spillover effect and control risks can get satisfactory returns and will help improve the price discovery function of A-share market.