零膨胀模型在心肌缺血节段数影响因素研究中的应用

被引：20

作者：

曾平

刘桂芬

曹红艳

机构：

[1] 山西医科大学卫生统计教研室

来源：

中国卫生统计 | 2008年 / 05期

关键词：

零膨胀模型; 计数资料; 零过多; Poisson回归; 负二项回归;

D O I：

暂无

中图分类号：

R195 [卫生调查与统计];

学科分类号：

100401 ;

摘要：

目的讨论零膨胀模型在计数资料中的应用。方法应用零膨胀负二项模型分析冠心病患者心肌缺血节段数的影响因素。结果零膨胀负二项模型logit部分结果表明没有家族史、年龄越小、左室收缩末容积越小的患者发生心肌节段缺血的可能性较小;负二项部分结果表明有糖尿病史、有冠心病家族史、左室收缩末容积越大的患者发生心肌缺血节段数越多。结论当计数资料存在过多零计数时,应用零过多模型能够得到很好的拟合效果。

引用

页码：464 / 466

页数：3

共 9 条

[1] The relationship between truck accidents and geomet-ric de-sign of road sections:poisson versus negative binomial regression. Miaou SP. Accid Anal Prevent . 1994
[2] A demonstration of modeling count data with an application to physical activity. Slymen,DJ,Ayala,GX,Arredondo,EM,Elder,JP. Epidemiol Perspect Innov . 2006
[3] Modeling accident frequencies as ze-ro-altered probability processes:an empirical inquiry. Shankar V,Milton J,Mannering F. Accid.Anal Prev . 1997
[4] Regression Models for Categorical and Li mited Dependent-Variables. Long JS. . 1997
[5] Likelihood ratio tests for model selection and non-nested hypotheses. Vuong,Q.H. Econometrica . 1989
[6] Zero-inflated Poisson regression,withan application to defects in manufacturing. LAMBERT D. Technometrics . 1992
[7] Zero-inflated Poisson modeling to evaluate occupational safetyinterventions. Carrivicka PJ W,Lee AH,Yau KKW. Safety Science . 2003
[8] Cameron,AC,Trivedi,PK. Regression analysis of count data . 1998
[9] Analysis of longitudal data. Peter Diggle,Patrick Heagerty,Liang KY,et al. . 2001

← 1 →