Design And Implementation Of Multifactor Strategy Optimization System Based On Genetic Algorithm

With the development of computer technology and financial engineering theory,quantitative investment model is increasingly recognized by the investment community.Multi-factor investment strategy is a mature theory,and the years of investment practice test mature quantitative investment strategy.How to determine the weight of the model factor in the process of combining,especially in real time dynamic adjustment of the weight optimization process? Due to the large number of factors,the weight adjustment dimension is high,and the number of alternatives is large.In order to achieve the optimization of real-time performance,we must have efficient bionic algorithm to solve the problem of multi-factor weight optimization.Therefore,the design of multi-factor strategy optimization system based on genetic algorithm plays an important role in financial engineering theory to explore and the practical effect of investment practice.The financial market is a very complex information environment with a wide variety of noise and effective information.Financial investment needs to discriminate effective information and invest in a large amount of information.Therefore,how to use computer to replace manpower for model optimization becomes a very important issue in financial quantitative investment.In order to solve the strategy optimization in the stock market with an increasingly large amount of information,this paper combines the computer technology with the investment strategy and introduces an automatic optimization algorithm to design the multi-factor weight of the genetic algorithm with the investment index Sharp ratio as the objective function.Automatically optimize the stock picking system.By comparing traditional workflows,system verification is performed using training set and test set data.At the same time,in order to overcome the shortcomings that the genetic algorithm brings to the system,K-means clustering method is added to achieve more scientific and reasonable classification in the process of factor classification.By discussing the setting of the K-means algorithm,the examples verify that the system incorporating the K-means algorithm can further improve the system's robustness and feasibility.Finally,under the combination of MATLAB,R and MySql platforms,a fully automated multi-factor stock picking system is implemented.Compared with the traditional multi-factor stock picking system,it has the ability of automatic optimization and can generate multiple multi-factor stock selection strategies to meet demand on demand.