A Low-frequency Trading Strategy Development And Empirical Analysis Combining Granwell's Law And Inertial Strategies

The quantitative trading strategy appeared firstly in the stock market and then became a popular trading method in the futures market.The quantitative trading strategy is more or less commonly used in the developed foreign market while it is just a starting step in the Chinese market.The rigorous trading strategy helps the traders to avoid the influence during the trade,which comes from the human error or the psychological fluctuation,and it is welcome by the traders because of its preciseness and probability.With the development of the electronic technology,the computer science,the telecom technology,the artificial intelligence and the big data,more and more advanced techniques and talents join the financial business so that the quantitative trading is at the same time under well developing.Quantitative trading is based on the historical data,which is used to transfer to the mathematical modeling.Then the modeling evolves into the algorithm and finally the trading strategy is realized.Investigate the trading strategy more in detail,we can say that the data is the object,the modeling is the kernel,the algorithm is the realization and the financial theory is the essence.What kind of financial theory is able to support the quantitative strategy for the long term? The traditional financial experts explained the stead state of the market while the behavioral financial experts are describing its transient state.Whatever in the developed securities market such as US,EU,JP or in the developing stock market like China,the abnormalities described by the behavioral financial theory exist and impact.The Chinese stock market does not have a long history and consists mainly of private investors.The natural man might be more suitable to describe the participators of Chinese stock market rather than the rational man and it has significant influence.If the market follows EMH,the trading strategy seems no space to survive.The abnormities which explained by the behavioral financial theory become the source of the trading strategy.The behavioral financial theory supports it.The under-reaction theory and the over-reaction theory are the twins of behavioral theory.The former one believes that the market participator has a lagged reaction to the information and the winning portfolio keeps good result in the following period;while the latter one agrees that there is an overshoot caused by the participator's reaction so that the winning portfolio performs too much at present and getting loss in the nearly future.Noise trading theory supports that the market contains the information trader and the noise trader.The noise exists in the market and significantly influents the price for the long term.In this paper a low frequency component momentum trading strategy,which was developed for the foreign exchange market by a researcher previously,and its empirical test are used for reference to the Chinese 300 ETF.After the empirical test it shows that the dual moving average filtered by the Hodrick-Prescott algorithm has positive effect to the 300 ETF.During the total data in-sample back test,the kernel algorithm of this strategy represents a suitable trend of the selected security but the value of the main evaluation items differ from each other quite a lot by the different parameters.On the other hand during the out-of-sample back test,this strategy requires a relatively long period historical data.These two points have brought some limitations to the real application of this strategy.In this paper the key algorithm is kept and some elements of the noise trading theory are added,the noise trader is defined according to the Granville rules whose suggestion was to buy or sell when the price was far away from the average value.Do the empirical test again and the annual return getting more stable according to the different parameters during in-sample back test and not such long historical data is requested during out-of-sample back test.Each trading strategy is developed to win in the market.But every single trade brings the tax and fee.Whatever the trades make benefit or get loss,the tax and fee are unavoidable.The tax and fee are finally one of the profit eaters.During the development of the low frequency trading strategy,the influence of the tax and fee has been fully considered.This paper consists of five parts: the introduction,the behavioral financial theory,the development outline of the trading strategy,the empirical test of multiple trading strategies and the conclusion.The main contribution of this paper is to make the trading strategy more flexible by keeping the algorithm of the researcher and merging some new elements of the behavioral financial theory.Applying and empirical analyzing the new strategy to the 300 ETF makes also good performance.