|
|
Non-ignorable, non-monotone missing data |
|
Message/Author |
|
|
I have an advanced cancer quality of life dataset, where missing data is not only non-ignorable, but also non-monotone as at some time points subjects missed filling out a question but didn't drop out of the study. Is there any literature to describe how to model such missing data that is non-monotone in MPlus? Thank you! |
|
|
Both monotone and non-monotone missing data can be handled in Mplus with ML under the standard MAR assumption. Non-ignorable missing data, that is non-MAR (NMAR), is more typically modeled to account for dropout, that is monotone missingness. The assumption probably is that non-monotone missingness is more likely to be MAR and NMAR methods are mostly needed for monotone situations. I discussed NMAR modeling in Muthén, B., Asparouhov, T., Hunter, A. & Leuchter, A. (2011). Growth modeling with non-ignorable dropout: Alternative analyses of the STAR*D antidepressant trial. Psychological Methods, 16, 17-33. These models also take care of non-monotone missingness that is MAR. NMAR modeling for non-monotone missingness is also possible, say using pattern-mixture techniques (see paper). |
|
|
Hello, I'm trying to follow the Star*D analysis. I started with the run2 (http://www.statmodel.com/examples/stard/run2.out) syntax. But receive the error *** ERROR in DATA MISSING command The number of variables specified for NAMES must be equal to the number of BINARY variables for TYPE=MISSING. My code is Missing = all (-999); USEVAR ARE GH1 GH2 GH3 GH4 GH5 GH6 GH7 GH8 GH9 GH10 u2 u3 u4 u5 u6 u7 u8 u9 u10; Data missing: names = GH1 GH2 GH3 GH4 GH5 GH6 GH7 GH8 GH9 GH10; binary = u2 u3 u4 u5 u6 u7 u8 u9 u10; I can't find much help online for the data missing command. Any links or thoughts would be much appreciated. |
|
|
P. 530 of the UG version on our website describes this and agrees with the error message you got: you have 9 u's and 10 GHs but there should be 10 u's. |
|
Back to top |
|
|