基于PCA-GWR方法的村级贫困时空格局及致贫因素分析

doi:10.12082/dqxxkx.2020.190286

地球信息科学学报 ›› 2020, Vol. 22 ›› Issue (2): 231-245.doi: 10.12082/dqxxkx.2020.190286

• 地理空间分析综合应用 • 上一篇下一篇

基于PCA-GWR方法的村级贫困时空格局及致贫因素分析

罗耀文¹, 任周鹏², 葛咏^2,^*(), 韩李涛¹, 刘梦晓², 何亚文³

^1. 山东科技大学测绘科学与工程学院,青岛 266590
^2. 中国科学院地理科学与资源研究所资源与环境信息系统国家重点实验室,北京 100101
^3. 中国石油大学（华东）,青岛 266580

收稿日期:2019-06-10 修回日期:2019-11-22 出版日期:2020-02-25 发布日期:2020-04-13
作者简介:罗耀文（1994— ）,女,山东淄博人,硕士生,研究方向为贫困时空分析。E-mail: luoyw@lreis.ac.cn
基金资助:
国家杰出青年基金项目(41725006);山东省自然科学基金项目(ZR2017MD003)

Analysis on Spatio-temporal Patterns and Drivers of Poverty at Village Level based on PCA-GWR

LUO Yaowen¹, REN Zhoupeng², GE Yong^2,^*(), HAN Litao¹, LIU Mengxiao², HE Yawen³

^1. College of Geomatics, Shandong University of Science and Technology, Qingdao 266590, China
^2. State Key Laboratory of Resources and Environmental Information System, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
^3. China University of Petroleum, Qingdao 266580, China

Received:2019-06-10 Revised:2019-11-22 Online:2020-02-25 Published:2020-04-13
Contact: GE Yong
Supported by:
The National Science Fund for Distinguished Young Scholars(41725006);Natural Science Foundation of Shandong Province(ZR2017MD003)

摘要/Abstract

摘要：

探究贫困的时空变化及识别致贫因素,可以为扶贫政策的制定和实施提供参考。贫困是由多种因素造成的,地理加权回归（GWR）可以分析各因素对贫困的影响在空间上的差异,但致贫因素之间存在较强的相关性会导致多重共线性问题。本文探索了基于主成分的地理加权回归方法（PCA-GWR）,结合自然、经济和社会属性对贫困空间格局特征进行因素分析;为探究贫困的时空变化规律,探索用全局Moran's I指数、局部G系数对村级贫困发生率的时空格局变化特征进行分析。并以江西省永新县为研究区为实验区进行分析。研究结果表明：① PCA-GWR模型中变量的方差膨胀因子（VIF）值明显低于GWR模型变量的VIF值,PCA-GWR模型有效地解决了GWR模型中存在的多重共线性问题;② 永新县贫困格局分布与地形、植被分布等自然因素和低学历、缺乏劳动力、疾病等乡村主体自生发展能力相关,且每种影响因素与贫困发生率的关系呈现出不同的空间模式;③ 2013—2017年永新县贫困发生率从11.27%下降至0.97%,呈现出逐年下降趋势,且村间贫困差距逐年缩小,其中2013—2015年贫困发生率分布西高东低,2016年和2017年整体值较低;④ 从空间相关性来看：全局上,2013—2016年表现出空间正相关,2017年呈现随机分布;局部上,2013—2016年的冷、热点分布变化不大,冷点分布在中部,热点聚集在西南部,2017年热点分布在南部,冷点零星分布于北部地区。研究结果可为政府扶贫政策的制定提供参考。

关键词: PCA-GWR模型, 多重共线性, 主成分分析, 致贫因素, 时空格局, 永新县

Abstract:

Exploring the spatio-temporal changes of poverty and identifying the factors that cause poverty can provide reference for the formulation and implementation of poverty alleviation policies.Poverty is caused by many factors. Geographically Weighted Regression model (GWR) can analyze the spatial differences in the influence of various factors on poverty,but there is a strong correlation between the factors causing poverty,which leadsto multicollinearity. Principal Component-based Geographic Weighted Regression method (PCA-GWR) is usedin this paper by combining the natural, economic and social attributes toanalyze the characteristics of the spatial pattern of poverty.In order to explore the spatio-temporal changes of poverty, this paper analyzes the temporal and spatial patterns of village-level poverty incidence from 2013 to 2017. Spatial autocorrelation analysis was performed using global Moran's I index and local G coefficientrespectively.Selecting Yongxin County of Jiangxi Province as the research area, the results show that: (1) There is a high correlation between independent variables affecting poverty. When these variables are put together in GWR model, the multicollinearity problem is easy to occur, and the results of GWRanalysis are not reliable. In order to eliminate the multicollinearity problem, Principal Component Analysis (PCA) was performed on the variables that were significantly correlated with the dependent variables. Three principal components were extracted by principal component analysis, including self-development ability of rural subjects, topographic and vegetation index. The Variance Inflation Factors（VIF）value of the variable in the PCA-GWR model is significantly lower than that in the GWR model. The PCA-GWR model effectively solves the multicollinearity problem in the GWR model. (2) The result of PCA-GWR found that the poverty in Yongxin County is the result of the combination of natural factors such as topographic factors and vegetation distribution and the self-development ability of rural subjects such as low-education, lack of labor, disease. And the effects of these factors presented different spatial patterns. This can provide a reference for the formulation of government poverty alleviation policies. (3) From 2013 to 2017, the incidence of poverty in Yongxin County decreased from 11.27% to 0.97%, showing a downward trend year by year, and the poverty gap between villages decreased year by year. The incidence of poverty from 2013 to 2015 was high in the west and low in the east. The overall value in 2016 and 2017 was low. (4) From the perspective of spatial correlation: on the whole, the spatial correlation between 2013 and 2016 is positive, and it is randomly distributed in 2017; Locally, the distribution of cold and hot spots did not change much from 2013 to 2016, the cold spots were distributed in the middle, and the hot spots were concentrated in the southwest. In 2017, hot spots are distributed in the south, and cold spots are scattered in the north.

Key words: PCA-GWR model, multicollinearity, principal component analysis, causes of poverty, spatio-temporal pattern, Yongxin County

罗耀文, 任周鹏, 葛咏, 韩李涛, 刘梦晓, 何亚文. 基于PCA-GWR方法的村级贫困时空格局及致贫因素分析[J]. 地球信息科学学报, 2020, 22(2): 231-245.DOI:10.12082/dqxxkx.2020.190286

LUO Yaowen, REN Zhoupeng, GE Yong, HAN Litao, LIU Mengxiao, HE Yawen. Analysis on Spatio-temporal Patterns and Drivers of Poverty at Village Level based on PCA-GWR[J]. Journal of Geo-information Science, 2020, 22(2): 231-245.DOI:10.12082/dqxxkx.2020.190286

图/表 15

图1

图2

表1

表2

表3

表4

图3

图4

表5

表6

图5

表7

图6

图7

表8

参考文献 33

[1]	联合国2030可持续发展目标[EB/OL]. https://sustainabledevelopment.un.org/?menu=130.
	[ United Nations 2030 sustainable development goals[EB/OL]. https://sustainabledevelopment.un.org/?menu=130. ]
[2]	丁建军, 冷志明 . 区域贫困的地理学分析[J]. 地理学报, 2018,73(2):232-247.
	[ Ding J J, Leng Z M . Regional poverty analysis in a view of geography science[J]. Acta Geographica Sinica, 2018,73(2):232-247. ]
[3]	刘艳华, 徐勇 . 中国农村多维贫困地理识别及类型划分[J]. 地理学报, 2015,70(6):993-1007.
	[ Liu Y H, Xu Y . Geographical identification and classification ofmulti-dimensional poverty in rural China[J]. Acta Geographica Sinica, 2015,70(6):993-1007. ]
[4]	王艳慧, 钱乐毅, 段福洲 . 县级多维贫困度量及其空间分布格局研究——以连片特困区扶贫重点县为例[J]. 地理科学, 2013,33(12):1489-1497.
	[ Wang Y H, Qian L Y, Duan F Z . Study on county-level multidimensional poverty measurement and its spatial distribution pattern: A case study of poverty alleviation key counties in contiguous poverty-stricken areas[J]. Scientia Geograohica Sinica, 2013,33(12):1489-1497. ]
[5]	邹薇, 方迎风 . 怎样测度贫困:从单维到多维[J]. 国外社会科学, 2012,2(7):63-69.
	[ Zou W, Fang Y F . How to measure poverty:from single dimension to multi-dimension[J]. Foreign Social Sciences, 2012,2(7):63-69. ]
[6]	贾林瑞, 刘彦随, 刘继来 , 等. 中国集中连片特困地区贫困户致贫原因诊断及其帮扶需求分析[J]. 人文地理, 2018,33(1):85-93.
	[ Jia L R, Liu Y S, Liu J L , et al. Cause diagnosis and demand analysis of poverty-stricken households in contiguous poverty-stricken areas in China[J]. Human Geography, 2018,33(1):85-93. ]
[7]	李平 . 东北贫困区多维致贫因素评价及实证分析[J]. 统计与决策, 2019,35(7):93-96.
	[ Li P . Evaluation and empirical analysis of multidimensionalfactors contributing to poverty in northeast poverty-stricken areas[J]. Statistics and Decision-making, 2019,35(7):93-96. ]
[8]	张琦, 陈伟伟 . 连片特困地区扶贫开发成效多维动态评价分析研究——基于灰色关联分析法角度[J]. 云南民族大学学报(哲学社会科学版), 2015,32(1):136-142.
	[ Zhang Q, Chen W W . Multidimensional dynamic evaluation and analysis of poverty alleviation effect in contiguous poverty-stricken areas: Based on the perspective of gray correlation analysis[J]. Journal of Yunnan university for Nationalities (philosophy and social sciences edition), 2015,32(1):136-142. ]
[9]	毛婧瑶, 葛咏, 赵中秋 , 等. 武陵山贫困片区扶贫成效评价与空间格局分析[J]. 地球信息科学学报, 2016,18(3):334-342.
	[ Mao J Y, Ge Y, Zhao Z Q , et al. Poverty alleviation evaluation and spatial pattern analysis of Wuling Mountain[J]. Journal of Geo- information Science, 2016,18(3):334-342. ]
[10]	陈烨烽, 王艳慧, 王小林 . 中国贫困村测度与空间分布特征分析[J]. 地理研究, 2016,35(12):2298-2308.
	[ Ceng Y F, Wang Y H, Wang X L . Measurement and spatial analysis of poverty-stricken villages in China[J]. Geographical Research, 2016,35(12):2298-2308. ]
[11]	袁媛, 伍彬, 古叶恒 . 重庆市城市贫困空间特征和影响因素研究——兼论东西部城市的异同[J]. 人文地理, 2015,30(1):70-77.
	[ Yuan Y, Wu B, Gu Y H . Study on the spatial characteristics and influencing factors of urban poverty in Chongqing--on the similarities and differences between eastern and western cities[J]. Human Geography, 2015,30(1):70-77. ]
[12]	文琦, 施琳娜, 马彩虹等. 黄土高原村域多维贫困空间异质性研究——以宁夏彭阳县为例[J]. 地理学报, 2018,73(10):1850-1864.
	[ Wen Q, Shi L N, Ma C H , et al. Spatial heterogeneity of multidimensional poverty in villages on loess plateau: A case study of Pengyang county, Ningxia[J]. Acta Geo-graphica Sinica, 2008,73(10):1850-1864. ]
[13]	徐孝勇, 封莎 . 中国14个集中连片特困地区自我发展能力测算及时空演变分析[J]. 经济地理, 2017,37(11):151-160.
	[ Xu X Y, Feng S . Measurement of self-development capacity and spatio-temporal evolution of 14 contiguous poverty-stricken areas in China[J]. Economic Geography, 2015,30(1):70-77. ]
[14]	童新华, 梁俏, 韦燕飞 . 广西县域贫困空间格局及影响因素分析[J]. 农村经济与科技, 2018,29(13):1-6.
	[ Tong X H, Liang Q, Wei Y F . Spatial pattern and influencing factors of poverty in Guangxi counties[J]. Rural Economy And Technology, 2008,29(13):1-6. ]
[15]	王永明, 王美霞, 吴殿廷 , 等. 贵州省乡村贫困空间格局与形成机制分析[J]. 地理科学, 2017,37(2):217-227.
	[ Wang Y M, Wang M X, Wu D T , et al. Analysis of spatial pattern and formation mechanism of rural poverty in Guizhou province[J]. Scientia Geographica Sinica, 2017,37(2):217-227. ]
[16]	丁建军 . 中国11个集中连片特困区贫困程度比较研究——基于综合发展指数计算的视角[J]. 地理科学, 2014,34(12):1418-1427.
	[ Ding J J . Comparative study on poverty degree in 11 contiguous poverty-stricken areas in China: From the perspective of comprehensive development index calculation[J]. Scientia Geographica Sinica, 2014,34(12):1418-1427. ]
[17]	周蕾, 熊礼阳, 王一晴 , 等. 中国贫困县空间格局与地形的空间耦合关系[J]. 经济地理, 2017,37(10):157-166.
	[ Zhou L, Xiong L Y, Wang Y Q , et al. Comparative study on poverty degree in 11 contiguous poverty-stricken areas in China: From the perspective of comprehensive development index calculation[J]. Economic Geography, 2017,37(10):157-166. ]
[18]	迟瑶, 王艳慧, 房娜 . 连片特困区贫困县农村基本公共服务与县域经济时空格局演变关系研究[J]. 地理研究, 2016,35(7):1243-1258.
	[ Chi Y, Wan Y H, Fang N . Spatio-temporal pattern of rural basic public services and county economy in poverty-stricken contiguous destitute areas of China[J]. Geographical Research, 2016,35(7):1243-1258. ]
[19]	罗庆, 樊新生, 高更和 , 等. 秦巴山区贫困村的空间分布特征及其影响因素[J]. 经济地理, 2016,36(4):126-132.
	[ Luo Q, Fan X S, Gao G H , et al. Spatial distribution characteristics and influencing factors of poor villages in Qinba mountains[J]. Economic Geography, 2016,36(4):126-132. ]
[20]	黄国勇, 张敏, 夏咏 , 等. 新疆边境贫困县自然地理环境影响因素实证分析[J]. 干旱区地理, 2015,38(4):814-820.
	[ Huang G Y, Zhang M, Xia W , et al. An empirical analysis of the factors affecting the physical geographical environment of poverty-stricken counties in Xinjiang border areas[J]. Arid Land Geography, 2015,38(4):814-820. ]
[21]	陈烨烽, 王艳慧, 赵文吉 , 等. 中国贫困村致贫因素分析及贫困类型划分[J]. 地理学报, 2017,72(10):1827-1844.
	[ Chen Y F, Wang Y H, Zhao W J , et al. Analysis of factors contributing to poverty in poor villages in China and classification of poverty types[J]. Acta Geographica Sinica, 2017,72(10):1827-1844. ]
[22]	武鹏, 李同昇, 李卫民 . 县域农村贫困化空间分异及其影响因素——以陕西山阳县为例[J]. 地理研究, 2018,37(3):593-606.
	[ Wu P, Li T S, Li W M . Spatial differentiation of rural poverty in county area and its influencing factors: A case study of Shanyang County in Shanxi province[J]. Geography Research, 2018,37(3):593-606. ]
[23]	马雄威 . 线性回归方程中多重共线性诊断方法及其实证分析[J]. 华中农业大学学报(社会科学版), 2008(2):78-81,85.
	[ Ma X W . Diagnosis method and empirical analysis of multicollinearity in linear regression equation[J]. Journal of Huazhong Agricultural University (social science edition), 2008(2):78-81,85. ]
[24]	李平 . 东北贫困区多维致贫因素评价及实证分析[J]. 统计与决策, 2019,35(7):93-96.
	[ Li P . Evaluation and empirical analysis of multidimensional factors contributing to poverty in northeast poverty-stricken areas[J]. Statistics And Decision-making, 2019,35(7):93-96. ]
[25]	李严洁 . 多元回归中的多重共线性及其存在的后果[J]. 中国卫生统计, 1992(1):24-27.
	[ Li Y J . Multicollinearity in multiple regression and its consequences[J]. China Health Statistics, 1992(1):24-27. ]
[26]	Yu D, Peterson N A, Reid R J . Exploring the impact of non-normality on spatial non-stationarity in geographically weighted regression analyses: Tobacco outlet density in New Jersey[J]. Giscience & Remote Sensing, 2009,46(3):329-346.
[27]	Bollinger G . Book Review: Regression Diagnostics: Identifying Influential Data and Sources of Collinearity[J]. Journal of Quality Technology, 198148(6):1044-1044.
[28]	肖雪梦, 张应应 . 三种回归方法在消除多重共线性及预测结果的比较[J]. 统计与决策, 2015(24):75-78.
	[ Xiao X M, Zhang Y Y . Comparison of three regression methods in eliminating multicollinearity and prediction results[J]. Statistics And Decision Making, 2015(24):75-78. ]
[29]	谢小韦 . 浅析多元线性回归中多重共线性问题的三种解决方法[J]. 科技信息, 2009(28):117-118.
	[ Xie X W . Three solutions to multicollinearity problems in multiple linear regression[J]. Science and Technology Information, 2009(28):117-118. ]
[30]	张润楚 . 多元统计分析[M]. 北京: 科学出版社, 2006.
	[ Zhang R C. Multivariate statistical analysis[M]. Beijing: Science Press, 2006. ]
[31]	Tobler W R . A computer movie simulating urban growth in the Detroit region[J]. Economic geography, 1970,46(sup1):234-240.
[32]	张松林, 张昆 . 全局空间自相关Moran指数和G系数对比研究[J]. 中山大学学报(自然科学版), 2007,46(4):93-97.
	[ Zhang S L, Zhang K . Comparative study on global autocorrelation Moran index and G coefficient[J]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2007,46(4):93-97. ]
[33]	宋鸿, 陈晓玲 . 运用空间自相关分析中国入境旅游增长空间格局[J]. 世界地理研究, 2006(1):99-106.
	[ Song H, Chen X L . Spatial autocorrelation analysis of China's inbound tourism growth spatial pattern[J]. World Geography Research, 2006(1):99-106. ]

贫困维度	贫困指标	Pearson相关系数	相关性程度
自然因素	高程	0.289**	显著弱相关
	NDVI	0.282**	显著弱相关
	坡度	0.031	不显著
	坡度>8°面积占比	0.214**	显著弱相关
社会因素	疾病	0.711**	显著强相关
	到最近镇驻地距离	-0.083	不显著
	缺乏劳动力	0.887**	显著极强相关
	夜间灯光	-0.173**	显著极弱相关
	低学历	0.986**	显著极强相关
经济因素	人均收入	0.107	不显著

	疾病	缺乏劳动力	低学历	坡度>8°面积占比	高程	NDVI
疾病	1.000	0.786**	0.737**	0.149*	0.164*	0.159*
缺乏劳动力	0.786**	1.000	0.912**	0.229**	0.291**	0.297**
低学历	0.737**	0.912**	1.000	0.194**	0.268**	0.281**
坡度>8°面积占比	0.149*	0.229**	0.194**	1.000	0.686**	0.397**
高程	0.164*	0.291**	0.268**	0.686**	1.000	0.689**
NDVI	0.159*	0.297*	0.281*	0.397**	0.689**	1.000

主成分	特征值	贡献率%	累计贡献率%
第一主成分	3.127	52.124	52.124
第二主成分	1.701	28.352	80.476
第三主成分	0.605	10.088	90.564
第四主成分	0.278	4.634	95.198
第五主成分	0.206	3.428	98.626
第六主成分	0.082	1.374	100

原始变量	第一主成分	第二主成分	第三主成分
疾病	0.619	-0.005	0.094
缺乏劳动力	0.534	0.127	0.139
低学历	0.525	0.128	0.091
坡度>8°面积占比	0.055	0.136	1.232
高程	0.070	0.504	0.843
NDVI	0.078	0.728	0.235

VIF统计	低学历	疾病	缺乏劳动力	高程	NDVI	坡度>8°面积占比
VIF均值	4.94	2.53	6.05	3.16	2.02	2.01
VIF最大值	7.32	3.64	8.69	3.86	2.30	2.55
VIF最小值	3.66	2.08	4.51	2.58	1.78	1.55

基于PCA-GWR方法的村级贫困时空格局及致贫因素分析

Analysis on Spatio-temporal Patterns and Drivers of Poverty at Village Level based on PCA-GWR

RichHTML

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 33

相关文章 15

Metrics

本文评价

推荐阅读 0

VIF统计	第一主成分	第二主成分	第三主成分
VIF均值	1.02	1.03	1.03
VIF最大值	1.06	1.15	1.19
VIF最小值	1.00	1.00	1.00

年份	Moran's I	z值	p值
2013	0.343	8.878	0.001
2014	0.288	7.487	0.001
2015	0.347	8.737	0.001
2016	0.248	6.269	0.001
2017	-0.014	0.392	-0.308

[1]	杨雨, 宋福铁, 张杰. 金融网络中心性对中国城市经济增长的影响机理[J]. 地球信息科学学报, 2023, 25(5): 982-998.
[2]	马力, 潘竟虎. 中国荒野地空间识别及时空演变[J]. 地球信息科学学报, 2023, 25(2): 324-339.
[3]	刘乐, 盛科荣, 王传阳. 中国科技型初创企业的时空格局及影响因素研究——基于创业生态系统视角[J]. 地球信息科学学报, 2023, 25(2): 340-353.
[4]	张歆越, 高晓路, 柴琪, 宋敦江. 中国地级城市建成区蓝绿空间时空格局及其气候影响因素分析[J]. 地球信息科学学报, 2023, 25(1): 190-207.
[5]	黄盛, 李卫江, 朱梦茹, 刘振. 降雨事件对上海地铁通勤客流时空影响的精细尺度研究[J]. 地球信息科学学报, 2022, 24(2): 249-262.
[6]	张超鹏, 陈鹏, 江欢, 于越. X市入室盗窃犯罪人空间出行距离及其影响因素分析[J]. 地球信息科学学报, 2022, 24(10): 1957-1967.
[7]	许正森, 徐永明. 整合DMSP/OLS和NPP/VIIRS夜间灯光遥感数据的长江三角洲城市格局时空演化研究[J]. 地球信息科学学报, 2021, 23(5): 837-849.
[8]	张泽, 胡宝清, 丘海红, 邓雁菲. 桂西南喀斯特-北部湾海岸带生态环境脆弱性时空分异与驱动机制研究[J]. 地球信息科学学报, 2021, 23(3): 456-466.
[9]	柳林, 孙秋远, 肖露子, 宋广文, 陈建国. 涉毒人员日常活动对盗窃警情空间格局影响的时间差异[J]. 地球信息科学学报, 2021, 23(12): 2187-2200.
[10]	葛咏, 刘梦晓, 胡姗, 任周鹏. 时空统计学在贫困研究中的应用及展望[J]. 地球信息科学学报, 2021, 23(1): 58-74.
[11]	蒲东川, 王桂周, 张兆明, 牛雪峰, 何国金, 龙腾飞, 尹然宇, 江威, 孙嘉悦. 基于独立成分分析和随机森林算法的城镇用地提取研究[J]. 地球信息科学学报, 2020, 22(8): 1597-1606.
[12]	李玉, 李奕燃, 王光辉, 石雪. 基于加权指数函数模型的高光谱图像分类方法[J]. 地球信息科学学报, 2020, 22(8): 1642-1653.
[13]	郭文杰, 秦昆, 毛宗福, 王一喆, 汪瑶, 许刚, 马露. “一带一路”成员国居民健康状况的时空统计分析[J]. 地球信息科学学报, 2020, 22(5): 1133-1141.
[14]	董鹤松, 李仁杰, 李建明, 李帅. 基于DMSP-OLS与NPP-VIIRS整合数据的中国三大城市群城市空间扩展时空格局[J]. 地球信息科学学报, 2020, 22(5): 1161-1174.
[15]	刘纪远, 张增祥, 张树文, 颜长珍, 吴世新, 李仁东, 匡文慧, 史文娇, 黄麟, 宁佳, 董金玮. 中国土地利用变化遥感研究的回顾与展望——基于陈述彭学术思想的引领[J]. 地球信息科学学报, 2020, 22(4): 680-687.

年份	热点个数	冷点个数	不显著个数
2013	16	31	185
2014	19	33	180
2015	23	36	173
2016	28	26	178
2017	22	11	199