R help - Jul 2011 - Dealing with missing values in a linear mixed model

Hello,

?? I am trying to run a linear mixed model using 


model.a <- lme(Psstotals ~ Visit, data=caf, random= ~ Visit|Id)


My dataset looks lie the following:


? ? Id?? Visit? Agecorrected Psstotals
1 106???? 0?????????? 19??????????? 8???????? 
2 106???? 1?????????? 19??????????? 9 ??????? 
3 106???? 2?????????? 19?????????? NA?????? 
4 106???? 3?????????? 19?????????? NA?????? 
5 106???? 4?????????? 19?????????? NA?????? 
6 106???? 5?????????? 19?????????? 11 ????? 
........................
.............

I have 14 visits for each ID and I do have some missing values for each Id. I
get the following error when I run the above model:

Error in na.fail.default(list(Visit = c(0L, 1L, 2L, 3L, 4L, 5L, 6L, 7L,? : 
? missing values in object


I do not want to do a listwise deletion as I would like to run model with all
available data. So, I tried the following


model.a <- lme(Psstotals ~ Visit, data=caf, random= ~Visit |Id,
na.action=na.exclude)


Is this the right syntax?

Thanks,

Debs

Hi,
within lme(), I think it is only possible to do na.action = na.omit. The
default action (= na.fail) causes lme() to print an error message and
terminate if there are any incomplete observations

Best,
Christine

-----
-- 
Christine Adrion, Dipl.-Stat.,MPH

Ludwig-Maximilians-Universit?t M?nchen
IBE - Institut f?r Medizinische Informations-
verarbeitung, Biometrie und Epidemiologie 
Marchioninistr. 15
D- 81377 M?nchen
Germany
 
Tel:    +49 89 7095 7486
Fax:    +49 89 7095 7491
eMail:  adrion at ibe.med.uni-muenchen.de  
web:    http://www.ibe.med.uni-muenchen.de
--
View this message in context:
http://r.789695.n4.nabble.com/Dealing-with-missing-values-in-a-linear-mixed-model-tp3649747p3649941.html
Sent from the R help mailing list archive at Nabble.com.