Hi,
I think in such cases you need to use the 'wave' parameter (wave =
Date),
but I must admit I found contradictory comments on how missing values are
treated in geepack. Does someone know what geepack is doing with missing
values?
Popko
Giacomo Santini wrote:>
> Hi all
>
> I am analyzing a data set containing information about the behaviour of
> marine molluscs on a vertical wall. Since I have replicate observations
> on the same individuals I was thinking to use the geepack library.
>
> The data are organised in a dataframe with the following variables
>
> Date = date of sampling,
> Size = dimensions (mm)
> Activity duration of activity (min)
> Water = duration of splashing by waves
> Hgt = resting eight of each specimen before activity begin
> Individual = a code indicating the id of the specimen.
>
> I have up to 12 replicate observations for individual. Some observation
> are missing and I organized the data frame to have exactly 12 rows for
> each specimen, with NAs where there is a missing observation.
>
> The following model worked fine:
>
> gee1<-geese(Activity~Water, id=Individual, data=dataF, family=gaussian)
>
> but when I use other variables e.g
>
> gee2<-geese(Activity~Hgt+Size+Water, id=Individual, data=dataF,
> family=gaussian)
>
> I get the error message
>
> "Error in geese.fit(x, y, id, offset, soffset, w, waves, zsca, zcor,
> corp, :
> nrow(zsca) and length(y) not match"
>
> which I am not able to understand.
>
> The same problem has been reported in the list in 2006, but I have not
> found any response to it.
>
> Any suggestion?
>
>
> Giacomo
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>
--
View this message in context:
http://www.nabble.com/problem-with--geepack-tp15851828p21601979.html
Sent from the R help mailing list archive at Nabble.com.