Stock Market Decision Model Based On Intelligent Algorithm

The stock market is a nonlinear dynamic system full of all kinds of noise and is vulnerable to a variety of factors. Effective trading strategies are the cornerstone of successful trade. Both fundamental analysis and technical analysis are trying to select high-quality stocks through the analysis of market and the company. Through quantitative screening, analysis, and finding the objective stock selection model, quantitative stock selection theory in financial engineering aim to underlie driving factors and select stocks which have the potential to rise.Intelligent algorithms refer to a series of algorithms inspired by the wisdom of nature and of human wisdom. Genetic algorithm can quickly solve the problem of local optimization; Genetic programming, which is characteristic by its unique population structure, broaden the scope of application of genetic algorithm and can solve more general problems; Neural network algorithms have Several branches which can solve problems quickly and accurately; Fish swarm algorithm also has a variety of applications. These intelligent algorithms provide rich and efficient solutions for various problems in the financial engineering.This paper attempts to design a top-down stock selection system, and use a variety of intelligent algorithms to complete specific task either in optimization process or in decision-making process. Use historical data and previous research results to train this system and make it become a ’sensitive erudite thinker’. Therefore, this mechanical trading system can help us free of emotional impact in the decision-making process and investors can use a computer instead of to make their own decisions. In this paper, the stock selection model, which includes fundamental analysis and technical analysis, take the genetic algorithm as its core algorithm, and apply genetic programming, neural networks and fish swarm algorithm to specific process, in order to imitate the inference procedure of a mature investor. This model optimized previous models and is proved reliable by historical data, and can provide valuable information for A-share market investors.