摘要
二项分类logistic回归是医学研究中常用的方法，优势比及其95％可信区间是logistic回归分析最重要的参数值，直接反映了自变量作用的大小和方向。而样本含量、自变量筛选、变量赋值和结果解释则是进行logistic回归分析的关键问题，决定了回归分析是否能得到相对最佳的回归模型而准确反映自变量的影响作用，作者将就上述问题进行阐述。
Binary logistic regression analysis is frequently used in medical researches, odds ratio （OR） and 95% confidence index, which can directly measure the effect of independent variable on the responsing variables, which are the most important indicators of logistic regression analysis. Sample size, variable selection, variable assignment and result interpretation are the key steps to construct an optimal model, which can reflect the influence of explanatory variables correctly, and they will be discussed esDeciallv in this thesis
出处
《中国防痨杂志》
CAS
2016年第8期615-618,共4页
Chinese Journal of Antituberculosis
基金
中国疾病预防控制中心青年科研基金课题（2015A204、2016A201）
