论文部分内容阅读
利用柔性神经树模型的改进结构优化算法对影响股票市场的过程参数进行筛选,在精确度较高的前提下在比较短的时间内找到影响股票市场风险的重要参数。在柔性神经树模型的学习过程中,该算法的进化代数不是一个固定值,而是以误差率来控制进化代数,试验证明此算法使模型最优,效率和精确度非常高。柔性神经树模型的结构和参数优化分别由概率增强式程序进化和模拟退火算法完成。研究结果表明该改进方法对预测股票市场风险是非常有效的。
The improved structural optimization algorithm of flexible neural tree model is used to filter the process parameters that affect the stock market, and find the important parameters that affect the stock market risk in a relatively short period of time with high accuracy. In the course of studying the flexible neural tree, the evolutionary algebra of the algorithm is not a fixed value, but the evolution algebra is controlled by the error rate. Experiments show that this algorithm makes the model optimal, and its efficiency and precision are very high. The structure and parameter optimization of flexible neural tree model are respectively completed by probabilistic enhancement program and simulated annealing algorithm. The results show that the improved method is very effective in predicting the stock market risk.