The Research Of Dynamic Portfolio Optimization Model Based On Hybrid Intelligent Algorithm

Since 2004, the financial products is developing quickly,so the competitiveness of the financial institutions is focused on financial products.They have to plan an optimal portfolio for funds of financial products, to achieve the highest profit. At the same time, with the expansion of China’s securities market, the research and development of portfolio gradually improve and mature, but many studies are focused on the stock investment, research about the funds financial products is little.Genetic algorithm is based on the fittest survival biological evolved, the proposed algorithm has strong robustness, the ability of global search.so a lot of research use the genetic algorithm to solve the portfolio.but genetic algorithm is easy to be premature, and it is not stable and accurate enough to solve the problem of the high requirement of precision.This paper presents hybrid intelligent algorithm,mixing the ant colony optimization algorithm into the genetic algorithm. Using this hybrid intelligent algorithm to solve the portfolio of funds financial products.Firstly, analysis investment structurally based on the Markowitz mean-variance model.Then building a dynamic structured portfolio mode,considering the dynamic nature of fi nancial market.Secondly, using binary encoding with the weight of each asset value to code generate the initial population.Obtained preliminary combination solution by selection, crossover and mutation.Finally add the combination solution into initial information of the ant colony optimization algorithm.Then ant search path iterative and update the pheromone so to get a stable optimal solution.This paper makes ICBC financial "win-win" 3 financial products in 2014 fortyninth A as an example for simulation.Simulation results show that the genetic ant colony hybrid intelligent algorithm is feasible and effective. To portfolio, it is not only more accurate, but also more stable.