人工神经网络在经济预测中的应用

The Application of Artificial Neural Network in Economic Forecast

关于人工神经网络（ＡＮＮ）在预测领域的应用，存在着不同的意见。一些人认为用ＡＮＮ预测比其他传统预测方法要好，也有一些人认为ＡＮＮ预测仅是一种被夸大了的时髦。本文通过实例应用与统计回归预测方法进行对比，结果表明ＡＮＮ预测方法是一种可行的，相当于非线性回归的方法，它对用线性回归方法预测效果较差的问题具有独特的优势。同时，也提供了实施、理解和估计ＡＮＮ的方法。

There are different opinions in forecast field about the application of Artificial Neural Network (ANN). Somepeople think ANN is better than traditional forecast method; Others think ANN is just a kind of exaggerated fashion. In this paper, compared with statistics regression forecast method through the application. The result indicates that ANN is a kind of feasible,method equikal of examples to non-linear regression method, which has special superiority to the linar regression method. In the meantime, we provide the method of carrying out, understanding and estimating ANN.