摘要： 介绍基因表达式程序设计方法的基本原理，针对股票指数分析与预测问题，在经典的GEP算法基础上，提出一种基于动态变异算子的改进的GEP算法——IGEP算法，动态变异算子随着进化代数和染色体所含基因数目不同而变化，从而加快了GEP的收敛速度和精确度，对算法进行了复杂度和收敛性分析。设计一种基于IGEP的股票指数分析与预测算法，数值实验结果表明，该算法优于经典GEP算法，具有较广泛的通用性。

关键词: 基因表达式程序设计, 复杂度分析, 收敛性分析, 股票指数预测

Abstract: This paper introduces the basic principle of Gene Expression Programming(GEP). An improved GEP algorithm called IGEP based on dynamic mutation operator which is changed with the gene number of the genome and the number of evolutionary generation is presented. The algorithm complexity of IGEP is given in the paper. Furthermore, IGEP is applied in the solution of prediction in stock-price index. The simulation results show that the model found by IGEP is more accurate than the one of classic GEP and which proves the IGEP can be widely used in many fields.

Key words: Gene Expression Programming(GEP), complexity analysis, convergence analysis, prediction in stock-price index

中图分类号: 

• TP18