主成分回归对脑卒中发病与环境因素间关系的解析

脑卒中的诱发已经被证实与环境因素包括气温和湿度之间存在密切关系.对脑卒中的发病环境因素进行分析可以对脑卒中高危人群进行风险评估并及时采取干预措施,而平均气压、最高气压、最低气压、平均温度、最高温度、最低温度、平均相对湿度、最低相对湿度8个自变量之间的共线性使得用多元线性回归方法得到的回归方程的精度降低.运用主成分回归分析,对脑卒中发病人数与环境因素进行了深入解析,结合统计软件SPSS的分析结果,给出了计算主成分的正确表达式,并将主成分与发病人数进行多元线性回归,最终确定了脑卒中发病人数与8个环境因素间的数学模型.

Principal Component Regression In-depth Analysis on Stroke Incidence and Environmental Factors

Evoked brain stroke has been confirmed with environmental factors,including the existence of a close relationship between temperature and humidity. The incidence of environmental factors on stroke analysis to evaluate the risk of disease,can also be on stroke in high-risk groups to intervene timely. While the average pressure,maximum pressure,minimum pressure,average temperature,maximum temperature,minimum temperature,average minimum relative humidity relative humidity between 8 variables,serious collinearity makes using the regression equation of multiple linear regression method to get the accuracy greatly reduced. Principal component regression analysis( principal component analysis and multiple linear regression combined) is an improved regression method,can eliminate the adverse effects brought by multiple correlation regression model. Using this method,the stroke incidence and environmental factors of in-depth analysis,combined with the statistical analysis software SPSS,the correct expression for calculating the principal component is given,to overcome the many false and misleading the principal components analysis using SPSS software,textbooks and published articles. Then the principal components with the incidence of multiple linear regression,and ultimately determine the number of mathematical models of stroke and 8 environmental factors.