| 基于改进神经网络的GDP时间序列预测 |
| |
| 引用本文： | 梁娜,张吉刚.基于改进神经网络的GDP时间序列预测[J].河南科学,2011,29(12):1506-1508. |
| |
| 作者姓名： | 梁娜 张吉刚 |
| |
| 作者单位： | 咸宁学院,湖北 咸宁,437100 |
| |
| 基金项目： | 国家自然科学基金(11171102);湖北省高等学校青年教师深入企业行动计划项目(XD20100659) |
| |
| 摘 要： | 由于GDP时间序列具有线性和非线性的特征,神经网络(NN)方法和集成预测方法等在预测分析时可能产生较大误差.以GDP的年增长率作为神经网络的输入,建立基于BPNN的GDP预测模型.利用此改进BPNN模型对我国的GDP进行预测和验证,并分别与ARIMA-BP集成模型及BPNN模型进行比较.结果表明,改进的BPNN模型预测...
|
| 关 键 词： | BP神经网络 GDP预测 准确率 |
Research of the GDP Prediction Based on the Improved BP Neural Network |
| |
| Liang Na
,
Zhang Jigang.Research of the GDP Prediction Based on the Improved BP Neural Network[J].Henan Science,2011,29(12):1506-1508. |
| |
| Authors: | Liang Na Zhang Jigang |
| |
| Institution: | (Department of Mathematics,Xianning College,Xianning 437100,Hubei China) |
| |
| Abstract: | Because the time series of the GDP hawe both linear and nonlinear characteristics,traditional forecasting methods,such as neural network and some integrated model,tend to bring errors.The proposed method,BP neural network model using the annual incremental rate of GDP for the network input,was set up to predict the GDP.The GDP forecastting results from improved BPNN model were compared with ARIMA-BP and single BPNN model,showing more accuracy. |
| |
| Keywords: | BP neural network GDP prediction accuracy rate |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
