Pairs Trading Studying In Chinese Stock Market

In 2015,Chinese stock market was very irregular,and there were many "monster" stocks.Investors need a strategy to bring a stable income because their income can't be guaranteed.At present,quantitative strategy has been widely used in the European and American markets because of its ability to gain a stable income.Among many quantitative strategies,pairs trading based on co-integration theory is mostly well known.The strategy considers that there is a phenomenon of stock price mean reversion between two stocks with co-integration.Specifically,the price change of two stocks with co-integration is random,but their spread have a reasonable interval,which market will correct this trend when the spread is out of the range,and let it return to the reasonable range.Pairs trading can be divided into three steps: choose stock pool,screening stock pool,set trading signals.The first step is going to narrow the range of options to reduce the work;the second step is going to determine the pairs trading object;the third step is going to determine how to trade the object.In the previous studies,scholars mainly concentrated in the third step,where the scholars just select the same industry stocks without further subdivision.According to Wind industry classification standard,this paper is going to extract four kinds of stocks from the Shanghai and Shenzhen 300 Index constituent stocks which are in industry,optional consumer,energy and information technology.With reference to D Harhoff's view mentioned in a paper published in 1996(the view which upstream enterprises generally have the key technology or key resources,do not contact with the consumer,downstream enterprises have the final products,and contact with the consumer).Taking two industry trading stocks as example,I will trade all stocks in the same industry,upstream and downstream enterprises with daily datas,30 m high-frequency datas and60 m high-frequency datas,using the average yield and matching success rate as evaluation indexes.Finally,the effect of trading with upstream enterprises and downstream enterprises are better than the same industry(take one upstream enterprise stock and one downstream enterprise stock to trade)no matter what kind of data frequency.At last I draw two conclusions:firstly when we begin to trade with the same industry stocks,if we can continue to subdivide them into upstream and downstream enterprises,we can indeedly improve the pair trading revenue situation,secondly increasing the data frequency does not necessarily improve pair trading revenue situation,scholars or practitioners should consider carefully when they choose the data frequency.