| With the continuous development of China's capital market,investors have increasing choices of financial derivatives and investment strategies.Since 2010,margin trading and stock index futures successively have been proposed,making it possible for short selling on the domestic market.Under the circumstance that up to now,there are 950 shares related with margin trading business,statistical arbitrage,as a low-risk but high-yield investment strategy,must be favored in the capital market.Statistical arbitrage is a strategy which is based on a market-neutral investment.When the relative price of assets deviates from the theoretical value while the relative prices return to a balanced level,reverse operation gains profit by building asset bull(undervalued)and bears(overvalued).This policy is suitable for the market to be short and the development of margin trading business,greatly boosting the investor's interest.This thesis selected the bank stocks with heavy weighting in market capitalisation of the a-share market for pairs trading and used the date of 5-minute stock and 60-minute closing price combined with margin trading business to built linear cointegration model.Taking into account the case where the sequence of model residuals might be fluctuating with time,the author used AR-GARCH model fitting residue.On the basis of trade signal,the author gained the optimal trading strategy in different closing prices as well as the simulated income in the sample and the outside forecast earnings comparative analysis.According to the empirical analysis in this thesis,statistical arbitrage in domestic capital market has great development space.A comparison of earnings for 5 minutes and 60 minutes found that high-frequency trading data had more arbitrage opportunities.In the forecast period,the model reflected sensitivity of time,which can build a better fitting model for the residual so that the statistical arbitrage strategy allows more investors to benefit. |
PDF Full Text Request