基于分位回归兰州市PM_(2.5)影响因素分析

The Influencing Factors on PM_(2.5) Concentration of Lanzhou Based on Quantile Regression

利用兰州市空气质量监测数据和气象记录数据构建线性分位数回归模型对比分析二氧化硫、二氧化氮、风速、气温、相对湿度对位于不同条件分位点的PM_(2.5)浓度的影响大小。构建半参数可加分位数回归模型分析各影响因素对位于不同条件分位点的PM_(2.5)浓度影响的非线性迹象。结果表明,各个分位点上,各因素对PM_(2.5)浓度的影响存在很大差异,而且有明显的不同的非线性迹象。总体来看,同种影响因素对不同浓度水平的PM_(2.5)影响存在很大差异。

According to air quality test data and meteorological record data of Lanzhou,a linear regression model is built to analyze the difference of some effect factors,including sulfur dioxide,nitrogen dioxide,wind speed,temperature,and relative humidity,on the concentration of PM2.5 at different contional points and a additive semiparametric regression model was constructed to analyze the nonlinear effects of each influencing factor on PM2.5 concentrations at different conditional points.The results show that there are significant differences in the effects of the factors on the PM2.5 concentrations at each point,and there are obvious non-linear signs.Overall,there are significant differences in the effects of the same factors on PM2.5 at different conditional points.