【摘要】 以《湖南省2015—2020年贫困村名单》为研究样本,通过探索性回归分析选取自然及社会经济等10个解释变量,结合空间自相关和核密度分析对县域贫困村空间分异特征进行分析,利用全局回归(OLS)模型和地理加权回归(GWR)模型对县域贫困村影响因素的空间异质性进行研究。结果表明:(1)湖南省贫困村分布具有显著的空间聚集性,常德—益阳—长沙—湘潭—衡阳—郴州西部边界以西是贫困村的主要聚集区域,边界以东贫困村明显偏少,东北、东南方向有少量聚集区;(2)GWR模型校正R2(0.831)高于OLS模型的校正R~2(0.719),且拟合优度较高;(3)10个解释变量与贫困村数量关系呈现不同数值大小、不同正负效应的空间分异规律。研究结果为合理制定产业发展策略、优化要素投入、缩小县域发展差异提供了参考。更多还原

【Abstract】 Based on the "List of Poor Villages in Hunan Province from 2015 to 2020", this paper selected 10 explanatory variables such as natural and social economy through spatial exploratory analysis, and combines spatial autocorrelation and nuclear density analysis methods to analyze the characteristics and spatial differentiation characteristics of poverty-stricken villages in counties. The spatial heterogeneity of the influencing factors of poverty-stricken villages in the county was studied by using the global regression(OLS) model and the geographically weighted regression(GWR) model. The distribution of poverty-stricken villages has significant spatial clustering in Hunan. The Changde-Yiyang-Changsha-XiangtanHengyang-Chenzhou of western boundary is the main gathering area of poverty-stricken villages, and the poverty-stricken villages in the east are obviously less, a small number of gathering areas in the northeast and southeast was found. The GWR model correction R2(0.831) was higher than the correction R~2(0.719) of the OLS model, and the goodness of fit was better. The relationship between the 10 explanatory variables and the number of poor villages showed the spatial differentiation law of different numerical values and different positive and negative effects. The research results provided a reference for rational formulation of industrial development strategies, optimization of factor inputs, and narrowing of county development differences.更多还原