部分线性模型在贵州省公路货运量预测中的应用
|
摘要
运用部分线性模型对贵州省公路货运量进行预测研究.首先运用灰色关联度分析法确定影响贵州省公路货运量的主要影响因子;然后运用主成分分析法将选取的影响因子指标数据进行降维处理,通过分析处理后的数据得到部分线性模型;最后,以2010-2012年的公路货运量作为验证值,将部分线性模型、多元线性回归模型及灰色预测模型的预测结果进行比较.研究结果表明:部分线性模型能较好地拟合贵州省1990-2009年公路货运量;三种模型的预测结果显示,部分线性模型预测结果优于多元线性回归模型和灰色预测模型的预测结果.
The highway freight forecasting of Guizhou province based on partial linear model is investigated in this paper. Firstly, the main impact factors of highway freight are obtained by grey correlation analysis method. Secondly, the related data which affects the highway freight of Guizhou is reduced by the principal component analysis method, and then the partial linear model is derived by analyzing the dimensional data. Finally, the Guizhou highway freight from2010 to 2012 is taken as verified values and the predictive results among partial linear model,multiple regression model and grey prediction model are compared. It is shown that the partial linear model fit well the highway freight in Guizhou by the data from 1990 to 2009;the empirical test proves that the proposed forecasting method based on partial linear model is high accuracy than the multiple regression model and grey prediction model.
The highway freight forecasting of Guizhou province based on partial linear model is investigated in this paper. Firstly, the main impact factors of highway freight are obtained by grey correlation analysis method. Secondly, the related data which affects the highway freight of Guizhou is reduced by the principal component analysis method, and then the partial linear model is derived by analyzing the dimensional data. Finally, the Guizhou highway freight from2010 to 2012 is taken as verified values and the predictive results among partial linear model,multiple regression model and grey prediction model are compared. It is shown that the partial linear model fit well the highway freight in Guizhou by the data from 1990 to 2009;the empirical test proves that the proposed forecasting method based on partial linear model is high accuracy than the multiple regression model and grey prediction model.
引文
[1]贵州省统计局.2014年贵州交通运输保持较快增长[EB/OL].http://www.gz.stats.gov. cn/Web62/News/20150304/5967. htm.
[2]赵建有,周孙锋,崔晓娟等.基于模糊线性回归模型的公路货运量预测方法[J].交通运输工程学报,2012,12(3):80-85.
[3]徐照.基于多元回归模型预测区域货运量一以无锡市市区货运量预测为例[J].企业改革与管理,2014,22:39.
[4]王栋.基于神经网络算法的公路货运量预测方法[J].北华大学学报(自然科学版),2014, 15(3):417-420.
[5]盖春英,裴玉龙.公路货运量灰色模型-马尔可夫链预测方法研究[J].中国公路学报,2003, 16(3):113-116.
[6] Engk R F, Granger C W J, and R ice J. et al. Semiparametric estimates between weather and electricity sales[J]. Journal of American Statistical Association, 1986, 80:310-319.
[7] Schamlensee R, Stoker T M.Household gasoline demand in the United States[J]. Econometrica,1999, 67:645-662.
[8]黄四民,梁华.用半参数部分线性模型分析居民消费结构[J].数量经济技术经济研究,1994, 10:33-38.
[9]孙志华,刘金犯,贾超华.部分线性模型在北京市税收分析与预测中的应用[J].数学的实践与认知,2011,41(4):9-13.
[10]刘锋,银利,张星.部分线性模型在空气质量指数细颗粒物PM2.5中的分析应用[J].数学的实践与认识,2014, 44(9):130-134.
[2]赵建有,周孙锋,崔晓娟等.基于模糊线性回归模型的公路货运量预测方法[J].交通运输工程学报,2012,12(3):80-85.
[3]徐照.基于多元回归模型预测区域货运量一以无锡市市区货运量预测为例[J].企业改革与管理,2014,22:39.
[4]王栋.基于神经网络算法的公路货运量预测方法[J].北华大学学报(自然科学版),2014, 15(3):417-420.
[5]盖春英,裴玉龙.公路货运量灰色模型-马尔可夫链预测方法研究[J].中国公路学报,2003, 16(3):113-116.
[6] Engk R F, Granger C W J, and R ice J. et al. Semiparametric estimates between weather and electricity sales[J]. Journal of American Statistical Association, 1986, 80:310-319.
[7] Schamlensee R, Stoker T M.Household gasoline demand in the United States[J]. Econometrica,1999, 67:645-662.
[8]黄四民,梁华.用半参数部分线性模型分析居民消费结构[J].数量经济技术经济研究,1994, 10:33-38.
[9]孙志华,刘金犯,贾超华.部分线性模型在北京市税收分析与预测中的应用[J].数学的实践与认知,2011,41(4):9-13.
[10]刘锋,银利,张星.部分线性模型在空气质量指数细颗粒物PM2.5中的分析应用[J].数学的实践与认识,2014, 44(9):130-134.