[R] Missing panels in multipanel lme lattice/trellis

Dieter Menne dieter.menne at menne-biomed.de
Fri Nov 23 08:55:34 CET 2001


> I don't have Pinheiro/Bates handy;  would you be able to send me the code
> that produces the output so I can take a look?

Here is a the example, with a simulated missing block/variety.
Following Kjetil's suggestion, I manually added a Skip-list.
Maybe a more general way to create the skip-list would be good 
enough, but possible one needs a dummy-panel that is inserted
for missing matrix-combinations.

# Trellis with missing graphs
# Remove one set to simulate missing data
Oats1<-Oats[(Oats$Block!="III") | (Oats$Variety!="Victory"),]
# attempted correction, stupid, but good enough for ONE plot
SkipIt[7]<-T # just "manual"
fm4Oats<-lme( yield ~ nitro, data = Oats1, random = ~ 1 | Block/Variety )
# produces Figure 1.21 of Pinheiro/Bates, 
# with one missing. 
plot( augPred( fm4Oats ), aspect = 2.5, layout = c(6, 3),
      between = list( x = c(0, 0, 0.5) ),skip=SkipIt ) 
# The plot is not really nice, I would prefer an empty
# box instead of the hole

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

More information about the R-help mailing list