首页 | 本学科首页   官方微博 | 高级检索  
检索        

倾向指数匹配法与Logistic回归分析方法对比研究
引用本文:王永吉,蔡宏伟,夏结来,蒋志伟,王陵.倾向指数匹配法与Logistic回归分析方法对比研究[J].现代预防医学,2011,38(12):2217-2219.
作者姓名:王永吉  蔡宏伟  夏结来  蒋志伟  王陵
作者单位:1. 第四军医大学预防医学系卫生统计学教研室,西安,710032
2. 第四军医大学口腔医学院信息中心
基金项目:国家自然科学基金资助课题,基于HIS的多分组临床观察数据倾向指数匹配的医学循证研究(30800952)
摘    要:目的]比较倾向指数匹配法与Logistic回归分析方法的检验效能和I类错误。方法]通过Monte Carlo模拟比较倾向指数匹配法和Logistic回归处理二分类资料的区别。结果]倾向指数匹配法和Logistic回归的I类错误无差异,倾向指数全匹配法的检验效能略高。结论]在观察性研究中,倾向指数匹配法具有很高的实用价值。

关 键 词:倾向指数  匹配  Logistic回归  Monte  Carlo模拟

COMPARISON OF PROPENSITY SCORE MATCHING AND LOGISTIC REGRESSION
WANG Yong-ji,CAI Hong-wei,XIA Jie-lai,et al..COMPARISON OF PROPENSITY SCORE MATCHING AND LOGISTIC REGRESSION[J].Modern Preventive Medicine,2011,38(12):2217-2219.
Authors:WANG Yong-ji  CAI Hong-wei  XIA Jie-lai  
Institution:WANG Yong-ji,CAI Hong-wei,XIA Jie-lai,et al.(The Fourth Military Medical University,Xi'an 710032,China)
Abstract:Objective]To compare the statistical power and Type I error rates between propensity score and Logistic regression.Methods]Monte Carlo simulations were adopted to compare the propensity score matching and Logistic regression when they were used to analyze nominal data.Results]There was no difference between propensity score matching and Logistic regression in Type I error rates.The nearest available neighbor matching had higher power than other methods.Conclusion]Propensity score matching should be reco...
Keywords:Propensity score  Matching  Logistic regression  Monte Carlo simulation  
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号