基于多层回归分析城市雾霾成因模型

Model of Causes for Urban Fog-Haze Based on Multiple Regression Analysis

基于灰色系统理论,运用多层回归理论及数据挖掘技术,建立了雾霾强度与气象效应和非气象效应的多层混合效应模型,以避开雾霾形成的复杂过程,得到了雾霾与主要成因的量化关系.对模型的残差,得出了模型残差近似服从标准正态分布,表明所建模型具有较好的准确性.

Based on the grey system theory with multiple regression theory and data mining, multiple mixed effect model was established. It was fog-haze intensity with the meteorological effect and the non-meteorological effect, which could avoid the complex process of fog-haze formation, and get quantitative correlation between fog-haze with its main causes. The residual obeyed standard normal distribution similarly. The model established can reflect the actual problem well.