Improvement Of China's Stock Market Selection Model Based On Single-layer Filter

Since the opening of China's capital market,it has experienced the initial stage of the market in the 1990s when insider trading was rampant,market manipulation was frequent,and the index skyrocketed.Since the beginning of the 21st century,China's stock market has rapidly improved its position in the national economy.On October 31,2017,the total market value of the stock market as a share of GDP jumped to 273.43%;whereas the highest peak is in 2015,as much as 471.88%.Although the total market value of the stock market has increased linearly in the past decade,the Chinese ordinary investors and some institutional investors who participated in it,have not only failed to share the rapid development of China's economy,but suffered from a great loss.This paper attempts to carry out two consecutive upgrades based on the single-layer filter selection model of single factor,in order to establish a set of investment system with strong logic that can survive in the Chinese stock market for a long time,and provide some reference ideas for investors to participate in Chinese stock market.This paper attempts to effectively integrate subjective investment with quantitative investment,and fully exerts subjective initiative in terms of factor range and factor category,and combines the previous research results and the relevant experience accumulated by the author in investment practice,to firstly classify the core factors affecting stock price fluctuation between the definitions of the value class and the growth class,then on the basis of which three core stock selection factors are specified within each category.Combined with the characteristics of the stock selection factors and the unique characteristics of the Chinese stock market,one of the three stock selection factors is excluded,and within the final value category and the growth category each retain two core stock selection factors.For the four stock selection factors retained,the value class and growth type single-layer filter stock selection model were established according to the category of the factor,and the Shanghai and Shenzhen 300 was selected as the performance comparison benchmark,relying on the market-wide financial data and market data between May 01,2015 to October 31,2017,then the effectiveness of the stock selection model established by the above factors were tested one by one,and the stock selection effectiveness of the single-layer filter stock selection model relative performance benchmark was determined.After verifying the validity of the single-layer filter stock selection model,the two groups of value classes and growth groups were formed,with a total of four groups of stock selection models.On this basis,we try to upgrade the single-layer filter stock selection model to form a two-tier filter selection model with first-value-then-growth and first-growth-then-value.We first compare the efficiency of the double-layer filter stock selection model with each factor as the first factor and the stock selection performance of the single-layer filter stock selection model with the factor as the core stock selection factor,to verify the same factor as the first layer of the double-layer filter selection model of the layer filter is effective as a single-layer filter stock selection model with the factor as the core stock selection factor.In the further analysis of the data,we compared the system performance of the four-group double-layer filter stock selection model with value preference,and selected a group of the best performing stock selection models.The optimal model of the double-layer model is compared with the optimal model in the one-factor stock selection model consisting value factor.Under the same factor category,the double-layer filter stock selection model is compared to the single-layer filter model to verify the superiority of the performance of the double-layer stock selection model.In the process of data analysis of all models of single-layer filter and double-layer filter,we found that in the system effectiveness evaluation system,the net high-point fallback index is less discriminating,because the stock selection model we established mainly emphasizes stocks.The choice did not consider the timing operation,but in China's highly fluctuating capital market,the long-term survival,especially the long-term management of investors' funds,extremely demands for the net value fallback.Combining the Dow Theory and the Golan average line rule,we select the moving averages for a specific time period to perform timing operations,and compare the system performance after the timing operation with the system performance before timing.