论文部分内容阅读
目的分析2012年全国31个省的宏观因素,包括经济水平、交通、社会保障及卫生水平,对艾滋病发病数影响的空间变化特征。方法采用地理加权泊松回归模型拟合艾滋病发病数与宏观因素之间的关系,利用局部加权极大似然方法估计回归系数曲面,揭示各因素对发病数影响的空间异质性,并且与经典的泊松回归模型的估计结果进行对比分析。结果地理加权泊松回归模型能够将数据的地理位置信息有效的纳入回归分析框架,充分探索回归关系的空间非平稳性,其拟合效果明显优于经典的泊松回归模型。中国各地区的艾滋病发病数与经济水平、交通、社会保障及卫生水平这4个宏观因素密切相关。结论 4个宏观因素对艾滋病发病数的影响具有显著的空间异质性特征,因此需要依据各地区社会宏观因素的差异性制定因地制宜的艾滋病防控政策。
Objective To analyze the macroeconomic factors of 31 provinces in China in 2012, including the spatial variation characteristics of economic level, traffic, social security and health level and the impact on AIDS incidence. Methods Geographic weighted Poisson regression model was used to fit the relationship between the incidence of AIDS and macroscopic factors. The local weighted maximum likelihood method was used to estimate the regression coefficient surface to reveal the spatial heterogeneity of the impact of various factors on the incidence. Poisson regression model of the estimated results for comparative analysis. Results Geographically weighted Poisson regression model can effectively integrate the geographic location information of the data into the regression analysis framework and fully explore the spatial nonstationarity of the regression relationship, and its fitting effect is obviously better than the classical Poisson regression model. The incidence of AIDS in all regions of China is closely related to the four macroeconomic factors of economic level, transportation, social security and health level. Conclusion The four macroscopic factors have significant spatial heterogeneity on the incidence of AIDS. Therefore, it is necessary to formulate a local HIV / AIDS prevention and control policy based on the differences of social macroscopical factors in different regions.