"Yang, Richard" <dyang at nrcan.gc.ca> writes:
> I have a 10 x 423 data frame which consisting of response, time, subject,
> site, plot and covariates (continueous and categorical) measured at the
plot
> level. When the data frame was converted into a groupedData object, a
> warning appeared
>
> > A <- groupedData(ht ~ time | Subject, data = tt, outer = ~ site *
plot,
> + labels=list(y = "Height", x = "Time", units =
list( y = "(m)", x > "(Yr)")) )
> Warning message:
> argument lengths differ in: split(x, f).
I think you are missing a closing parenthesis in that call to
groupedData.
It should be
A <- groupedData(ht ~ time | Subject, data = tt, outer = ~ site * plot,
labels=list(y = "Height", x = "Time"),
units = list( y = "(m)", x ="(Yr)"))
That won't affect the fitting routines but it can affect the plots.
Please remember that you must use
library(lattice)
to do any of the plotting in the nlme package. Once R-1.5.0 is out
the nlme package will require the lattice package.
> The warning did not affect the estimation of a nlme model with parameters
> B1, B2, and B3. However, when plotting the estimated ranef versus
> covariates, an error resulted
>
> > A1Mod.nlmeRE <- ranef(A1Mod.nlme, aug=T)
>
> > plot(A1Mod.nlmeRE, form = B1 ~ Ps + Dr + El + As + Sl) results in
> Error in max(length(x0), length(x1), length(y0), length(y1)) :
> Argument "x0" is missing, with no default
>
> The same data frame and commands works fine with Splus 6 on W2K. Is this a
> bug in nlme 3.3.x? I use R 1.4.1.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._