Re: HELP WITH MISSING OBS IN REPEATED MEASURE DESIGN

["Lucke, Joseph F" <LUCKE@uthscsa.edu>]

This appears to be a case of informative missingness---The attribute you are measuring, tumor size, is the cause of the missingness.  Therefore standard MAR methods are not applicable. There is a rather large literature on this problem. I prefer pattern-mixture modeling for these sorts of data.  For a start, see

 

Fitzmaurice, G. M., Laird, N. M., & Shneyer, L. (2001). An alternative parameterization of the general linear mixture model for longitudinal data with non-ignorable drop-outs. Statistics in Medicine, 20, 1009--1021.

Hedeker, D., & Gibbons, R. D. (1997). Application of random-effects pattern-mixture models for missing data in longitudinal studies. Psychological Methods, 2(1), 64--78.

Hogan, J. W., & Laird, N. M. (1997). Mixture models for the joint distribution of repeated measures and event times. Statistics in Medicine, 16, 239--257.

Molenberghs, G., & Verbeke, G. (2001). A review on linear mixed models for longitudinal data, possibly subject to dropout. Statistical Modeling, 1, 235--269.

Verbeke, G., & Molenberghs, G. (2000). Linear mixed models for longitudinal data. New York: Springer.

 

Joe

 

 -----Original Message-----
From: s-news-owner@lists.biostat.wustl.edu [mailto:s-news-owner@lists.biostat.wustl.edu] On Behalf Of duo wan
Sent: Wednesday, May 12, 2004 10:38 PM
To: s-news@lists.biostat.wustl.edu
Subject: [S] HELP WITH MISSING OBS IN REPEATED MEASURE DESIGN

Hi,

 

I have a data set consisting of 3 treatment groups and 1 control group (10 subjects per group); at each of 8 time points, tumor sizes were measured. However, subjects in control group were dying at last 3 time points and none survived in the last time points. Since there are no observations for control group in the last time point, is it reasonable to drop all the observations in the last time point or do some kind of imputation?  The main interest is to find out if 3 treatments can reduce tumor size.

 

Any suggestion will be appreciated.

 

Vincent wan

 

 

---

Do you Yahoo!?
Yahoo! Movies - Buy advance 
tickets for 'Shrek 2'