摘要
基于DMSP/OLS夜间灯光数据和土地利用数据,对云南沿边地区国内生产总值(GDP)进行空间化及预测研究。结果表明:基于土地利用数据的第一产业模型的拟合优度(R 2)每期均大于0.82,拟合的相对误差均低于1.12%;基于夜间灯光数据,通过分类回归方法的第二、三产业模型每期的相对误差均低于7%;最终GDP每期拟合误差均低于4.3%。基于1992-2013年空间化的GDP数据,对2015年的GDP进行预测精度分析,对2020年的GDP进行预测。2015年预测的GDP除了福贡县等县市的相对误差大于25%外,其余各县各年份的相对误差均小于25%。预测2020年云南沿边地区GDP可达到5233.409亿元。
Based on the DMSP/OLS nighttime light data and land use data,a spatial study of the gross domestic product(GDP)of Yunnan border area was conducted.First,GDP was divided into the primary industry and the second,third industries.Then,the primary industry was modeled using land use data,and the DMSP/OLS nighttime light data was used to apply the“classification regression”method to model the second,third industries.Finally,the 22-year high precision spatial fitting model of GDP in Yunnan border area from 1992 to 2013 was realized.The results show that the goodness of fit(R 2)of the primary industry model is greater than 0.82 per period,and the relative error of the fitting is less than 1.12%.The relative error of the secondary and tertiary industry after the“classification regression”method is less than 7%per period.The final fitting error of GDP in the Yunnan border area is less than 4.3%.Based on 22-year of spatialized GDP data,the prediction accuracy of GDP in 2015 is analyzed and the GDP in 2020 is forecasted.The forecasting results show that the relative errors of the forecasted GDP in 2015,except for Fugong County and others are more than 25%and the relative errors of the remaining counties in each year are less than 25%.It is predicted that GDP of Yunnan border area can reach 523.3409 billion yuan by 2020.
作者
卢秀
李佳
段平
程峰
王金亮
LU Xiu;LI Jia;DUAN Ping;CHENG Feng;WANG Jinliang(College of Tourism and Geographical Sciences,Yunnan Normal University,Kunming 650500,China;Key Laboratory of Virtual Geographic Environment and Ministry of Education,Nanjing Normal University,Nanjing 210023,China)
出处
《地域研究与开发》
CSSCI
CSCD
北大核心
2020年第2期36-39,81,共5页
Areal Research and Development
基金
国家重点研发计划项目(SQ2018YFE011725)
国家自然科学基金项目(41561048)。
