期刊文献+

利用logit模型判定数据缺失机制 被引量:3

Using logit model to test the data missing mechanism
下载PDF
导出
摘要 利用logit模型刻画了缺失指示变量R的分布,由其分布的参数估计来判定数据的缺失机制类型.在四个假定的基础上,用五个步骤具体操作缺失数据的机制检验.并用两个例子说明了检验的具体步骤. Logit model was used to depict the distribution of the missing indicator variable. The data missing mechanism was tested according to its parameter estimations. On the basis of four assumptions, this paper uses five steps were used to operate the test of the missing mechanism. Two examples were illustred to show the details of the test.
作者 孙晓松
出处 《山东理工大学学报(自然科学版)》 CAS 2009年第2期51-54,共4页 Journal of Shandong University of Technology:Natural Science Edition
关键词 数据缺失机制 LOGIT模型 缺失指示变量 数据非随机缺失机制 data missing mechanism logit model missing indicator variable not missing at random mechanism
  • 相关文献

参考文献8

  • 1Rubin D B. Inference and missing data[J]. Biometrika, 1976,63: 581-592.
  • 2Cox D R. The analyss of binary data[M]. London, Chapman and Hall, 1970.
  • 3Fitzmaurice G M, Laird N M. A likelihood-based method for analysing longitudinal binary responses [J]. Biometrika, 1993, 80:141-151.
  • 4Conaway M R. Non-ignorable non-response models for time-ordered categorical variables[J]. Applied Statistics,1993, 42(1): 105-115.
  • 5Diggle P J, Kenward M G. Informative dropout in longitudinal data analysis[J]. Applied Statistics, 1994,43 : 49-93.
  • 6Baker S G. Marginal regression for repeated binary data with outcome subject to non-ignorable non-response[J] . Biometrics, 1995, 51(3): 1 042-1 052.
  • 7Troxel A B, Lipsitz S R, Harrington D P. Marginal models for the analysis of longitudinal measurements with non-ignorable non monotone missing data[J]. Biometrika, 1998, 85:661-672.
  • 8Wu L. Exact and approximate inferences for nonlinear mixedeffects models with missing covariates[J]. Journal of American Statistical Association, 2004,99 : 700-709.

同被引文献36

引证文献3

二级引证文献16

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部