论文部分内容阅读
根据中、美贸易的经济研究背景,分析中、美贸易重要影响指标,对其进行线性回归分析,发现中、美贸易额与各指标之间不是简单的线性相关,且各指标之间存在相互关系,所以不能用单一的线性回归模型进行预测。而BP神经网络具有非线性映射、自适应学习和良好的泛化能力等特征,运用BP神经网络模型对中、美贸易进行实证预测,大大提高了预测精度,取得了较好的效果。
According to the background of economic studies in China and the United States, analyzing the important influence indexes of China-US trade and conducting linear regression analysis on them, we find that the trade volume between China and the United States is not linearly correlated with each index and there is mutual Therefore, we can not predict with a single linear regression model. The BP neural network has such characteristics as non-linear mapping, adaptive learning and good generalization ability. The BP neural network model is used to predict the trade between China and the United States, which greatly improves the prediction accuracy and achieves good results.