文章基本信息

标题：基于BP_神经网络模型的延吉市GDP预测
其他标题：The Forecast of Yanji GDP Based on BP_Neural Network Model
本地全文：下载
作者：孙文渊 ; 尹哲 ; 沈京虎等
期刊名称：Business and Globalization
印刷版ISSN：2331-0189
电子版ISSN：2331-0219
出版年度：2015
卷号：03
期号：03
页码：65-73
DOI：10.12677/BGlo.2015.33008
出版社：Hans Publishers
摘要：经济预测问题是典型的多指标小样本复杂系统的预测问题。本文利用SAS软件做逐步回归分析法对原始数据进行预处理，使数据便于分析研究，预处理后利用仿真拟合、神经网络预测等方法根据2001~2009年延吉国民生产总值近九年的统计数据,研究出有关影响延吉市经济发展情况的显著影响因素X₁第一产业总值、X₄延吉市内游客数、X₅引入资金及X₆人口增长率。再进行独立性为目标的主成分分析并预测延吉市未来经济形式，对更好地掌握延吉市经济发展有一定参考价值。
其他摘要：Economic prediction problem is a complex system prediction problem of typical multi-index small sample. In this paper, by using SAS software with stepwise regression analysis method to do the preprocessing of the raw data, the data are easier to analyze. Using the methods of simulation fitting, neural network prediction and so on based on the statistical data of Yanji’s GDP from 2001 to 2009 after pretreatment, relevant factors which significantly affect economic development situation of Yanji city are worked out. Among them, X₁ is the total cost of the first industry; X₄ is the number of tourists in Yanji city; X₅ is the capital introduced and X₆ is the rate of population growth. Next we put the independence as the target of principal component analysis and predict Yanji city’s economic form in the future, having a certain reference value to better grasp Yanji city’s economic development.
关键词：GDP ; 逐步回归分析 ; 主成分分析 ; 神经网络模型 ; GDP ; Stepwise Regression Analysis ; Principal Component Analysis ; Neural Network Model
其他关键词：GDP ; 逐步回归分析 ; 主成分分析 ; 神经网络模型 ; GDP ; Stepwise Regression Analysis ; Principal Component Analysis ; Neural Network Model