Research Of The Efficiency Of The Stock Connect Under Analysis Of The Statistical Arbitrage Based On A+H Shares

In order to promote the opening-up of the capital market, to further promote the two-way open and healthy development of China’s capital market and the Hong Kong capital market, the authorities of the two-sides launched the Shanghai-Hong Kong Stock Connect program. The program means freer cross-border capital flows, opening up a channel between the A-share market and the overseas, it makes the mainland and Hong Kong investors to realize capital exchange.By the end of 2015, the program had been running for more than one year. The price spread between the A-and H-share has not been reduced, but expanded 40%. This will undoubtedly make people have doubts about the efficiency of the system.In this paper, we first analyze the impact of policies on the liquidity, profitability and volatility (or security) of A+H shares. Empirical results show that the liquidity of the A-share and H-share markets has been greatly improved after the Shanghai-Hong Kong Stock Connect program was approved, but has little effect on narrowing the difference between the liquidity of the A-share and H-share markets. After the program approved, although stock returns of the both sides have been well improved, the overall volatility has increased. Therefore, investors need to pay more attention to the risk of trading.Secondly, this paper tests the linkage effect between the price of A-and H-shares by means of correlation analysis, cointegration analysis and stability analysis. Empirical results show that the log-prices of the A+H pairs has a high correlation, so as the log-return series. The correlation has been greatly improved after the program approved and the stability of the price spread of A-and H-shares has been enhanced. But the cointegration relationship is not significant.Finally, we formulate a pair trading of "arbitrage portfolio strategy" to analyze the arbitrage space between A+H shares. Then we can further study the effect of policy on reducing the segmentation of the two sides and enhancing the linkage of the two markets.Because most of the cointegration relationship between A+H shares is not significant, we can’t use the traditional "cointegration strategy" pair trading.We use the high correlation between A- and H- shares to construct arbitrage portfolio. One is "1:1 pairing portfolio ", the other is "minimum variance portfolio ". We establish the ARMA-GARCH model directly on the stock return series, to scroll estimate and forecast the portfolio’s return and conditional variance.A moving time window method is used when testing the effect of out-of-sample pair trading. Specifically refers to the extrapolation time window length remained unchanged (180 days). The time window translates once a day, ARMA-GARCH equation, residual correlation coefficient, stock returns and conditional variance are estimated according to the new time window.The small changes over time of the regression parameters can be captured in this method to characterize the latest relevant relationship between A-and H-shares.In the aspect of signal selection, the "cointegration strategy" use the residual deviation as an arbitrage signal. In contrast, after forecasting the daily return and variance, we select the time-varying SHARP ratio as a trading signal to guide our trading. In the signal threshold selection, our strategy use the multiple of the standard deviation as open signal just as same as the general "cointegration strategy".Without considering the transaction cost and based on the assumption that the transaction has no friction, we simulate the pair trade of the 68 pairs of arbitrage portfolios and calculate the yield curve under "1:1 pairing strategy" and "MVP strategy". The empirical result indicates that the "1:1 pairing strategy" is better than the "MVP strategy"; Using time-varying SHARP ratio to guide the transaction can get a higher income than simply buying arbitrage portfolio. Under both two strategies, arbitrage opportunities and arbitrage gains have been reduced after the Shanghai-Hong Kong Stock Connect program approved, which proves that the program has a positive effect on reducing the market segmentation between the two markets.