摘要
提出了回归分析前作变量变换的依据 ,并认为变量的概率分布相同时可不作变量变换 ;应用江苏省土壤中镭和钍的数据建立的数学模型 ,对该省污染预测、重复调查本底具有现实意义。
The conditions for determining transformation on variables before regression analysis are proposed. Transformation on variables shouldn't be done when the probability distribution of variables is the same. The model which is put forword based on the Ra and Th data of soil in Jiangsu province has realistic significance for forecasting pollution and repetitively investigating background.
出处
《华东地质学院学报》
2002年第3期237-239,共3页
Journal of East China Geological Institute
关键词
回归分析
数学模型
辐射
环境监测
土壤
镭
钍
regression analysis
mathematical model
radiation
environmental monitoring