物流业发展的区位差异、驱动因素及时空异质性研究——基于GTWR模型的分析
|
摘要
基于2005—2015年中国省级面板数据,运用TOPSIS模型测度省区物流业发展水平,并结合核密度估计、探索性数据等方法研究物流产业演化规律,最终构建GTWR模型考察物流产业发展的驱动因素及其时空异质性。研究表明:中国省域物流业发展差异性与相关性并存;物流产业多维驱动要素呈现出明显的时空非平稳性,不同时点、地区各驱动要素的波动方向和强度并不相同,不同驱动要素分别呈现出左偏、右偏、对称、多峰等分布态势;各要素均呈现出一定的东、中、西梯度分布格局;不同地区各驱动要素的时变形态也存在不同。
Based on the Chinese provincial panel data from 2005 to 2015,this study utilizes TOPSIS model to measure the development level of provincial logistics industry,and combines Kernel density estimation method and ESDA method to analyze the lawof industrial evolution.Besides,the Geographical and Temporally Weighted Regression(GTWR) model is developed to explore the driving factors and spatial-temporal heterogeneity of the logistics industry.The results prove that spatial difference and spatial autocorrelation exist in the development of China's provincial logistics industry.The multi-dimensionally influencing factors of the logistics industry obviously show temporal and spatial non-stationarity.Moreover,the direction and intensity of the driving factors differ from time points and regions.From the perspective of the overall distribution of coefficients,different driving elements show the left,right,symmetrical,multi-peak and other distribution trends.However,from the viewof spatial difference of the coefficients,each element presents a certain gradient of east,middle and west pattern.Also,from the perspective of time series fluctuation of coefficients,the time-varying patterns of driving factors in different regions are different,mainly rising,falling and volatility.
Based on the Chinese provincial panel data from 2005 to 2015,this study utilizes TOPSIS model to measure the development level of provincial logistics industry,and combines Kernel density estimation method and ESDA method to analyze the lawof industrial evolution.Besides,the Geographical and Temporally Weighted Regression(GTWR) model is developed to explore the driving factors and spatial-temporal heterogeneity of the logistics industry.The results prove that spatial difference and spatial autocorrelation exist in the development of China's provincial logistics industry.The multi-dimensionally influencing factors of the logistics industry obviously show temporal and spatial non-stationarity.Moreover,the direction and intensity of the driving factors differ from time points and regions.From the perspective of the overall distribution of coefficients,different driving elements show the left,right,symmetrical,multi-peak and other distribution trends.However,from the viewof spatial difference of the coefficients,each element presents a certain gradient of east,middle and west pattern.Also,from the perspective of time series fluctuation of coefficients,the time-varying patterns of driving factors in different regions are different,mainly rising,falling and volatility.
引文
蔡海亚,徐盈之.2016.长江三角洲物流产业发展格局及影响机理研究:基于空间经济学的视角[J].华东经济管理(10):15-23.
陈恒,魏修建,杜勤.2015.“一带一路”物流业发展驱动因素的动态轨迹演变:基于劳动力投入的视角[J].上海财经大学学报(2):31-43.
韩兆洲,林仲源.2017.我国最低工资增长机制时空非平稳性测度研究[J].统计研究(6):38-51.
刘华军,何礼伟.2016.中国省际经济增长的空间关联网络结构:基于非线性Granger因果检验方法的再考察[J].财经研究(2):97-107.
李沙浪,雷明.2014.基于TOPSIS的省级低碳经济发展评价及其空间面板计量分析[J].中国管理科学(Special Issue):741-748.
吕光桦,宋文飞,李国平,等.2011.考虑空间相关性的我国区域研发全要素生产率测算:基于1999-2008年省际空间面板数据[J].科学学与科学技术管理(4):105-110,133.
马勇,黄智洵.2017.基于GWR模型的长江中游城市群传统村落空间格局及可达性探究[J].人文地理(4):78-85.
齐亚伟,陶长琪.2014.环境约束下要素集聚对区域创新能力的影响:基于GWR模型的实证分析[J].科研管理(9):17-24.
唐建荣,徐媛媛,李晓静.2015.基于SEM的城市物流驱动因素分析及对策研究:以江苏省为例[J].华东经济管理(8):39-46.
唐建荣,张鑫和.2017.物流业发展的时空演化、驱动因素及溢出效应研究:基于中国省域面板数据的空间计量分析[J].财贸研究(5):11-21.
王爱,陆林,包善驹,等.2017.基于GWR模型的合肥居住地价影响因素研究[J].人文地理(5):89-97.
王健,刘荷.2014.区域物流发展的影响因素研究:基于福建省的实证分析[J].华东经济管理(3):22-27.
魏修建,陈恒.2014.物流发展驱动要素对经济增长贡献度的区域差异性研究:基于丝绸之路经济带西北地区面板数据模型的实证分析[J].上海经济研究(6):14-22.
邬跃,梁晨,温卫娟.2007.外资进入中国物流业的影响[J].中国流通经济(5):18-20.
向书坚,许芳.2016.中国的城镇化和城乡收入差距[J].统计研究(4):64-70.
谢守红,蔡海亚.2015.长江三角洲物流业发展的时空演变及影响因素[J].世界地理研究(3):118-125.
姚娟,庄玉良.2012.外商直接投资对我国地区物流发展水平的影响研究[J].世界经济研究(7):60-66.
姚昕,潘是英,孙传旺.2017.城市规模、空间集聚与电力强度[J].经济研究(11):165-177.
张学良.2012.中国交通基础设施促进了区域经济增长吗:兼论交通基础设施的空间溢出效应[J].中国社会科学(3):60-77.
BAI Y,WU L X,QIN K,et al.2016.A geographically and temporally weighted regression model for ground-level PM2.5 estimation from satellite-derived 500m resolution AOD[J].Remote Sensing,8(3):262.
BRUNSDON C,AITKIN M,FOTHERINGHAM S,et al.1999.A comparison of random coefficient modelling and geographically weighted regression for spatially non-stationary regression problems[J].Geographical&Environmental Modelling,3(1):47-62.
CHU H J,HUANG B,LIN C Y.2015.Modeling the spatio-temporal heterogeneity in the PM10-PM2.5 relationship[J].Atmospheric Environment,102:176-182.
CLEVELAND W S,DEVLIN S J.1988.Locally weighted regression:an approach to regression analysis by local fitting[J].Journal of the American Statistical Association,83(403):596-610.
DINIZ-FILHO J A,SOARES T N,DE CAMPOS TELLES M P.2016.Geographically weighted regression as a generalized wombling to detect barriers to gene flow[J].Genetica,144(4):1-9.
FOTHERINGHAM A S,CHARLTON M,BRUNSDON C.1996.The geography of parameter space:an investigation of spatial non-stationarity[J].International Journal of Geographical Information Systems,10(5):605-627.
FOTHERINGHAM A S,CRESPO R,YAO J.2015.Geographical and temporal weighted regression(GTWR)[J].Geographical Analysis,47(4):431-452.
GUO Y,TANG Q,GONG D Y,et al.2017.Estimating ground-level PM2.5 concentrations in Beijing using a satellite-based geographically and temporally weighted regression model[J].Remote Sensing of Environment,198:140-149.
HWANG C L,YOON K.1981.Multiple attribute decision making[M].Heidelberg:Springer.
LIU J,ZHAO Y,YANG Y,et al.2017.A mixed geographically and temporally weighted regression:exploring spatial-temporal variations from global and local perspectives[J].Entropy,19(2):53.
PESSOA A.2010.R&D and economic growth:how strong is the link[J].Economics Letters,107(2):152-154.
RODRIGUEZ-POSE A.1999.Innovation prone and innovation averse societies:economic performance in Europe[J].Growth and Change,30(1):75-105.
SILVERMAN B W.1986.Density estimation for statistics and data analysis[M].London:Chapman&Hall.
TOBLER W R.1970.A computer movie simulating urban growth in the detroit region[J].Economic Geography,46(S1):234-240.
WU B,LI R,HUANG B.2014.A geographically and temporally weighted autoregressive model with application to housing prices[J].International Journal of Geographical Information systems,28(5):1186-1204.
(1)本文采用Gauss-Kruger Projection方法将观测点椭球体坐标系下的经纬度转为直角坐标系下的坐标。
(1)东部地区包括北京、天津、河北、辽宁、上海、江苏、浙江、福建、山东、广东、海南11个省区,中部地区包括黑龙江、吉林、山西、安徽、江西、河南、湖北、湖南8个省区,西部地区包括内蒙古、广西、重庆、四川、贵州、云南、西藏、陕西、甘肃、青海、宁夏、新疆12个省区。
(1)对空间位置进行随机排列,从而模拟计算出全局Moran指数的P值,即Pseodu p值。本文选择的排列次数(permutations)为999次。
(1)在GWR模型中,若带宽趋于无穷大,任意两点的权重将趋于1,则被估计的参数变成一致时,GWR就等于经典的OLS线性回归;反之当带宽变得很小时,参数估计将会更加依赖于邻近的观测值。
(2)GTWR模型综合考虑时间和空间信息进行参数估计。N×T的回归样本可得到N×T个系数,本文样本数量31×11,可得到341个估计系数。限于文章篇幅,各年不同地区的参数估计值不再一一列出,具体结果在图3中进行展示。
(1)创新悖论指创新不一定能有效地转化为增长(Pessoa et al.,2010)。
(2)社会过滤用于描述本地经济社会条件对创新转化为增长所产生的“过滤”作用,本地经济社会条件就好比创新与增长中间的“介质层”,“介质层”的不同导致创新转化率也不同(Rodriguez-pose et al.,1999)。
陈恒,魏修建,杜勤.2015.“一带一路”物流业发展驱动因素的动态轨迹演变:基于劳动力投入的视角[J].上海财经大学学报(2):31-43.
韩兆洲,林仲源.2017.我国最低工资增长机制时空非平稳性测度研究[J].统计研究(6):38-51.
刘华军,何礼伟.2016.中国省际经济增长的空间关联网络结构:基于非线性Granger因果检验方法的再考察[J].财经研究(2):97-107.
李沙浪,雷明.2014.基于TOPSIS的省级低碳经济发展评价及其空间面板计量分析[J].中国管理科学(Special Issue):741-748.
吕光桦,宋文飞,李国平,等.2011.考虑空间相关性的我国区域研发全要素生产率测算:基于1999-2008年省际空间面板数据[J].科学学与科学技术管理(4):105-110,133.
马勇,黄智洵.2017.基于GWR模型的长江中游城市群传统村落空间格局及可达性探究[J].人文地理(4):78-85.
齐亚伟,陶长琪.2014.环境约束下要素集聚对区域创新能力的影响:基于GWR模型的实证分析[J].科研管理(9):17-24.
唐建荣,徐媛媛,李晓静.2015.基于SEM的城市物流驱动因素分析及对策研究:以江苏省为例[J].华东经济管理(8):39-46.
唐建荣,张鑫和.2017.物流业发展的时空演化、驱动因素及溢出效应研究:基于中国省域面板数据的空间计量分析[J].财贸研究(5):11-21.
王爱,陆林,包善驹,等.2017.基于GWR模型的合肥居住地价影响因素研究[J].人文地理(5):89-97.
王健,刘荷.2014.区域物流发展的影响因素研究:基于福建省的实证分析[J].华东经济管理(3):22-27.
魏修建,陈恒.2014.物流发展驱动要素对经济增长贡献度的区域差异性研究:基于丝绸之路经济带西北地区面板数据模型的实证分析[J].上海经济研究(6):14-22.
邬跃,梁晨,温卫娟.2007.外资进入中国物流业的影响[J].中国流通经济(5):18-20.
向书坚,许芳.2016.中国的城镇化和城乡收入差距[J].统计研究(4):64-70.
谢守红,蔡海亚.2015.长江三角洲物流业发展的时空演变及影响因素[J].世界地理研究(3):118-125.
姚娟,庄玉良.2012.外商直接投资对我国地区物流发展水平的影响研究[J].世界经济研究(7):60-66.
姚昕,潘是英,孙传旺.2017.城市规模、空间集聚与电力强度[J].经济研究(11):165-177.
张学良.2012.中国交通基础设施促进了区域经济增长吗:兼论交通基础设施的空间溢出效应[J].中国社会科学(3):60-77.
BAI Y,WU L X,QIN K,et al.2016.A geographically and temporally weighted regression model for ground-level PM2.5 estimation from satellite-derived 500m resolution AOD[J].Remote Sensing,8(3):262.
BRUNSDON C,AITKIN M,FOTHERINGHAM S,et al.1999.A comparison of random coefficient modelling and geographically weighted regression for spatially non-stationary regression problems[J].Geographical&Environmental Modelling,3(1):47-62.
CHU H J,HUANG B,LIN C Y.2015.Modeling the spatio-temporal heterogeneity in the PM10-PM2.5 relationship[J].Atmospheric Environment,102:176-182.
CLEVELAND W S,DEVLIN S J.1988.Locally weighted regression:an approach to regression analysis by local fitting[J].Journal of the American Statistical Association,83(403):596-610.
DINIZ-FILHO J A,SOARES T N,DE CAMPOS TELLES M P.2016.Geographically weighted regression as a generalized wombling to detect barriers to gene flow[J].Genetica,144(4):1-9.
FOTHERINGHAM A S,CHARLTON M,BRUNSDON C.1996.The geography of parameter space:an investigation of spatial non-stationarity[J].International Journal of Geographical Information Systems,10(5):605-627.
FOTHERINGHAM A S,CRESPO R,YAO J.2015.Geographical and temporal weighted regression(GTWR)[J].Geographical Analysis,47(4):431-452.
GUO Y,TANG Q,GONG D Y,et al.2017.Estimating ground-level PM2.5 concentrations in Beijing using a satellite-based geographically and temporally weighted regression model[J].Remote Sensing of Environment,198:140-149.
HWANG C L,YOON K.1981.Multiple attribute decision making[M].Heidelberg:Springer.
LIU J,ZHAO Y,YANG Y,et al.2017.A mixed geographically and temporally weighted regression:exploring spatial-temporal variations from global and local perspectives[J].Entropy,19(2):53.
PESSOA A.2010.R&D and economic growth:how strong is the link[J].Economics Letters,107(2):152-154.
RODRIGUEZ-POSE A.1999.Innovation prone and innovation averse societies:economic performance in Europe[J].Growth and Change,30(1):75-105.
SILVERMAN B W.1986.Density estimation for statistics and data analysis[M].London:Chapman&Hall.
TOBLER W R.1970.A computer movie simulating urban growth in the detroit region[J].Economic Geography,46(S1):234-240.
WU B,LI R,HUANG B.2014.A geographically and temporally weighted autoregressive model with application to housing prices[J].International Journal of Geographical Information systems,28(5):1186-1204.
(1)本文采用Gauss-Kruger Projection方法将观测点椭球体坐标系下的经纬度转为直角坐标系下的坐标。
(1)东部地区包括北京、天津、河北、辽宁、上海、江苏、浙江、福建、山东、广东、海南11个省区,中部地区包括黑龙江、吉林、山西、安徽、江西、河南、湖北、湖南8个省区,西部地区包括内蒙古、广西、重庆、四川、贵州、云南、西藏、陕西、甘肃、青海、宁夏、新疆12个省区。
(1)对空间位置进行随机排列,从而模拟计算出全局Moran指数的P值,即Pseodu p值。本文选择的排列次数(permutations)为999次。
(1)在GWR模型中,若带宽趋于无穷大,任意两点的权重将趋于1,则被估计的参数变成一致时,GWR就等于经典的OLS线性回归;反之当带宽变得很小时,参数估计将会更加依赖于邻近的观测值。
(2)GTWR模型综合考虑时间和空间信息进行参数估计。N×T的回归样本可得到N×T个系数,本文样本数量31×11,可得到341个估计系数。限于文章篇幅,各年不同地区的参数估计值不再一一列出,具体结果在图3中进行展示。
(1)创新悖论指创新不一定能有效地转化为增长(Pessoa et al.,2010)。
(2)社会过滤用于描述本地经济社会条件对创新转化为增长所产生的“过滤”作用,本地经济社会条件就好比创新与增长中间的“介质层”,“介质层”的不同导致创新转化率也不同(Rodriguez-pose et al.,1999)。