| 基于DFA和SVR的GDP预测模型研究 |
| |
| 引用本文： | 张鹏.基于DFA和SVR的GDP预测模型研究[J].重庆文理学院学报,2015,34(5):42-45. |
| |
| 作者姓名： | 张鹏 |
| |
| 作者单位： | 太原工业学院理学系， 山西太原030008 |
| |
| 摘 要： | 首先对1980—2013年全国年度GDP数据进行消除趋势分析（DFA），发现GDP时序具有长记忆性，表明当前GDP值用来预测未来一段时间内的GDP值具有可行性，在此基础上利用具有非线性和强泛化能力特点的支持向量回归机（SVR），建立SVR预测模型,得到了较好的预测效果.
|
| 关 键 词： | 消除趋势分析 支持向量回归机 GDP |
Research of GDP prediction model based on DFA and SVR |
| |
| ZHANG Peng.Research of GDP prediction model based on DFA and SVR[J].Journal of Chongqing University of Arts and Sciences,2015,34(5):42-45. |
| |
| Authors: | ZHANG Peng |
| |
| Affiliation: | Science Department of Taiyuan Institute of Technology, Taiyuan Shanxi 030008, China |
| |
| Abstract: | This paper firstly uses the Detrended Fluctuation Analysis method (DFA) to analyze the national annual Gross Domestic Product (GDP) data from 1980 to 2013, which shows GDP time series has long memory, that is to say, using the current GDP data to predict the future GDP value over a period of time is practicable. Based on this and Support Vector Regression Machine (SVR) characterized by nonlinear and strong generalization, SVR prediction model is established and a better prediction effect is obtained in this paper. |
| |
| Keywords: | DFA SVR GDP |
|
| 点击此处可从《重庆文理学院学报》浏览原始摘要信息 |
|
点击此处可从《重庆文理学院学报》下载全文 |
|
