以蚊虫为例浅析不同气温参数的多重共线性对建模的影响

发布时间：2018-05-09 11:15

本文选题：气温 + 共线性　；参考：《中国媒介生物学及控制杂志》2015年04期

【摘要】：目的了解常用气温参数的多重共线性,为应用相关参数建立虫量的温度驱动消长模型提供参考。方法应用SPSS 13.0软件分析无锡市2010年3月至2011年11月每天的平均气温(X1)、最高气温(X2)、最低气温(X3)、18:00时气温(X4)、5日均温(X5)5个参数之间的相关性,并以本地蚊虫为例经向后法诊断了气温参数在蚊密度消长模型中的多重共线性。结果各参数之间呈直线相关,相关性均有统计学意义(P0.01)。其中X1与X2的相关性最高,校正决定系数(Rc2)=0.959;X4与X5相关性最低,Rc2=0.811。在以气温参数为自变量建立蚊密度消长模型时,随着自变量个数的增加,多重共线性程度也增强,模型的复相关系数(R)升高,Rc2降低。结论气温参数之间存在显著的自相关,在建模时存在较强的多重共线性,多重共线性实际上降低了回归方程的拟合度,在引入气温因子建立相关模型时应避免同时应用多个气温参数。
[Abstract]:Objective to study the multiple collinearity of common air temperature parameters, and to provide a reference for the establishment of temperature driven growth and decline model of insect mass by using related parameters. Methods SPSS 13.0 software was used to analyze the correlation among the five parameters of daily mean air temperature of Wuxi from March 2010 to November 2011, the maximum temperature was X _ 2 ~ (2 +), and the minimum temperature was X _ (3) ~ (1) ~ (18: 00). The multiple collinearity of temperature parameters in the model of mosquito density decline was diagnosed with local mosquitoes as an example. Results there was a linear correlation among the parameters, and the correlation was statistically significant (P 0.01). The correlation between X1 and X2 is the highest, and the correlation between X1 and X2 is the lowest. With the increase of the number of independent variables, the degree of multiple collinearity increases, and the complex correlation coefficient increases and Rc2 decreases when the temperature parameters are taken as independent variables. Conclusion there is a significant autocorrelation between air temperature parameters, and there is a strong multiple collinearity in modeling. In fact, multiple collinearity reduces the fitting degree of the regression equation. Multiple temperature parameters should be avoided when introducing air temperature factors to establish correlation model.
【作者单位】：无锡市疾病预防控制中心消毒与媒介生物防制科;
【基金】：无锡市卫生系统科研项目(Q201302)~~
【分类号】：R384.1

【参考文献】