论文部分内容阅读
文章基于四川省1978~2015年的国内生产总值(GDP)的经济增长数据,应用人工神经网络建立前5年的经济增长、第6年的相对时间与第6年的经济增长的关系模型,以实现用上5年的经济增长预测下1-5年的经济增长。研究结果表明,相对GDP的均方差函数mse为1.26091×10-5,相应的拟合精度为98%;全部预测年份(1983~2015)的平均预测精度为96%,表明预测值与实际值吻合程度很好,模型精度较高,建模简单。人工神经网络是预测区域经济增长的有效方法。
Based on the economic growth data of Sichuan Province from 1978 to 2015, the paper establishes the model of the relationship between the economic growth of the first 5 years and the 6th year of relative growth and the 6th year of economic growth by artificial neural network. In order to achieve the next five years of economic growth predicted next 1-5 years of economic growth. The results show that the mean square error of relative GDP, mse, is 1.26091 × 10-5, and the corresponding fitting accuracy is 98%. The average forecast accuracy of all forecast years (1983 ~ 2015) is 96%, indicating that the predicted value agrees with the actual value Very good, high precision model, modeling simple. Artificial neural network is an effective method to forecast the regional economic growth.