改进的实数遗传算法在求解组合预测模型中的应用

Using an improved real-code genetic arithmetic to solve weights of combination forecasting model

在现有文献的基础上,对实数遗传算法进行了改进,并将改进后的实数遗传算法应用到求解组合预测模型权重问题中。改进后的实数遗传算法模型预测精度有所提高,从而证明其能够较好地解决组合预测模型的权重求解问题。

On the base of present literatures, the paper improves the real-code arithmetic, and uses the improved method to solve weighs of combination forecasting model. The improved real-code genetic arithmetic enhances more precision of combination model than GA, which improves that the improved real-code genetic arithmetic can more preferable solve weights of combination forecasting model.