论文部分内容阅读
介绍了灰色预测模型GM(1,1)的结构和模型检验,以及BP神经网络预测模型的原理,对灰色预测模型进行改良,将此改良模型与BP神经网络模型进行组合,建立了新的组合灰色神经网络模型。以厦门市商品房成交量为例,以MATLAB为工具,进行2012年的成交量对比以及2013年成交量的预测,结果证明组合灰色神经网络的预测精度较高,可以为房地产价格指数预测研究提供参考依据。
The structure and model test of gray forecasting model GM (1,1) and the principle of BP neural network forecasting model are introduced. The gray forecasting model is improved. The improved model is combined with BP neural network model to establish a new combination Gray neural network model. Taking the volume of real estate in Xiamen as an example and the comparison between the volume in 2012 and the turnover in 2013 using MATLAB as a tool, the result shows that the forecast accuracy of the combined gray neural network is high, which can provide a reference for the real estate price index forecasting research in accordance with.