Trading Strategies Of China A Shares Based On Regular Reviews Of The CSI300Index

This paper builds up a model to predict the constituent changes of the CSI300index. Based on the predict model, the author develops a trading strategy to capturethe excess return of the changed constituent stocks during the regular reviews of theCSI300index every six months. Back test results using data from June2006toDecember2013confirms that the prediction accuracy ratio of the predict model isover90percent. By first predicting the constituent changes before the announcement,the portfolio that has long position in the stocks to be added to the index and shortposition in the stocks to be deleted from the index can make about3.9percent profitduring the holding period of four days before the announcement day of the constituentchanges until the implement day.The author further illustrates that the portfolio returns can be increased by about1percent if stocks with below-median arbitrage risk are excluded from the portfolio,which would be more attractive to risk-loving investors as the volatility also increases.Besides, the author also provides the return comparison between two weightingmethods and incorporates the trading cost. Using the standard event studymethodology the author proves that the strategy makes positive alpha in most cases,with a possibility over80percent. In the end, this paper discusses about theimplementation concerns and future potentiality of the strategy.