基于VAR模型的重庆市大气污染物分析及研究  

Analysis and Study of Air Pollutants in Chongqing Based on VAR Models

作　　者：安军[1,2] 王丹 AN Jun;WANG Dan(School of Mathematics and Statistics,Chongqing Technology and Business University,Chongqing 400067,China;Chongqing Key Laboratory of Social Economic and Applied Statistics,Chongqing Technology and Business University,Chongqing 400067,China)

机构地区：[1]重庆工商大学数学与统计学院,重庆400067 [2]重庆工商大学重庆市社会经济和应用统计重点实验室,重庆400067

出　　处：《重庆工商大学学报（自然科学版）》2023年第5期72-80,共9页Journal of Chongqing Technology and Business University:Natural Science Edition

基　　金：重庆工商大学重点平台开放项目(KFJJ2018099);重庆市自然科学基金面上项目(CSTC2020JCYJ-MSXMX0328);重庆市教委科学技术研究项目(KJQN202000838);重庆市社会科学规划重点委托项目(2020WT24);教育部人文社会科学研究规划基金项目(20YJA910002);国家社会科学基金后期资助一般项目(21FTJB002).

摘　　要：目的研究重庆市首要空气污染物PM_(2.5)与PM_(10)、SO_(2)、NO_(2)、CO、O_(3)的动态影响关系,为政府制定防治大气污染措施及相关政策提供有价值的建议。方法收集重庆市2021-05-01—2021-10-31日的PM_(2.5)、PM_(10)、SO_(2)、NO_(2)、CO、O_(3)这6项大气污染物的日浓度数据,利用Eviews8.0软件,对原始数据进行序列平稳性检验;根据Granger因果检验结果选择变量,建立时间序列VAR模型,并检验模型的稳定性;利用广义脉冲响应分析和方差分解分析,研究各污染物浓度对PM_(2.5)的动态影响及相对重要性。结果Granger因果检验表明:PM_(10)、SO_(2)、NO_(2)、O_(3)是PM_(2.5)的Granger原因,CO不是PM_(2.5)的Granger原因;广义脉冲响应分析表明:NO_(2)对PM_(2.5)的影响最大;方差分解分析表明:NO_(2)的浓度对PM_(2.5)的影响最大;O_(3)对PM_(2.5)的影响次之,对SO_(2)的影响作用最小。所以,从长期影响效应看,NO_(2)对PM_(2.5)具有长期较大的影响,SO_(2)对PM_(2.5)的影响最弱。结论防治PM_(2.5)对重庆市空气的污染应着重控制NO_(2)的污染,因此,政府应大力发展绿色交通,控制交通污染;大力监管高污染行业,将烟雾、粉尘、颗粒物等排放量较大的行业作为工业污染源治理的重点;大力发展清洁能源,加快化石燃料替代资源的开发利用。Objective The dynamic impact relationships between PM_(2.5)(the primary air pollutants in Chongqing)and PM_(10),SO_(2),NO_(2),CO and O_(3)were studied to provide valuable suggestions for the government to formulate air pollution prevention and control measures and related policies.Methods The daily concentration data of PM_(2.5),PM_(10),SO_(2),NO_(2),CO and O_(3)were collected from May 1,2021 to October 31,2021 in Chongqing,and the original data were tested for sequence stationarity using software of Eviews 8.0.Variables were selected based on the results of Granger causality tests,time series VAR models were developed and the stability of the models was tested.Generalized impulse response analysis and variance decomposition analysis were used to investigate the dynamic effects and relative importance of each pollutant concentration on PM_(2.5).Results The results of the Granger causality test showed that PM_(10),SO_(2),NO_(2)and O_(3)were the Granger causes of PM_(2.5),and CO was not the Granger cause of PM_(2.5).Generalized impulse response analysis showed that NO_(2)had the greatest effect on PM_(2.5).Variance decomposition analysis showed that the concentration of NO_(2)had the largest effect on PM_(2.5),O_(3)had the second largest effect on PM_(2.5)and SO_(2)had the smallest effect.Therefore,in terms of the long-term effect,NO_(2)has a large long-term effect on PM_(2.5),while SO_(2)has the weakest effect on PM_(2.5).Conclusion The prevention and control of PM_(2.5)pollution in the air in Chongqing should focus on controlling NO_(2)pollution.Therefore,the government should vigorously develop green traffic and control traffic pollution,vigorously supervise and regulate high-polluting industries and make industries with high emissions of smoke,dust,and particulate matter the focus of industrial pollution source governance,and vigorously develop clean energy and accelerate the development and utilization of alternative fossil fuel resources.

关 键 词：PM_(2.5) VAR模型 广义脉冲响应 方差分解分析 

分 类 号：F064.1[经济管理—政治经济学] X513[环境科学与工程—环境工程] X821