[100 Multilevel Questions]Missing at Random or Missing Not at Random

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

I am interested in the analysis of longitudinal data such as clinical trials with nonignorable dropout/missing.  I could find some references regarding the approaches of how to analyze data under both assumptions. However, little was found on how to test the missing is NMAR or MAR.   Would appreciate it if anyone can introduce me some reference on this topic.  Thanks.

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Multilevel questions question missing random interested references regarding clinical Random

Good morning

There also seems to be a consensus among the experts that if auxiliary' variables are added to the imputation model, in addition to the covariates in the analysis model, missingness may be better explained, and the MAR assumption may become more plausible

References

• Graham JW. Missing Data. Analysis and Design. Series: Statistics for Social and Behavioral Sciences.2012. Springer New York.
• van Buuren S, Flexible Imputation of Missing Data (2012), Chapman & Hall/CRC
• White, I.R., Royston, P., and Wood, A.M. (2011). Multiple imputation using chained equations: Issues and guidance for practice. Statistics in Medicine 30:377-399.
  

Richard A.Johnson《Applied Multivariate Statistical Analysis 》 习题答案