|
S-plus Users:
I am currently trying to create some conditioning plots using xyplot. I want to plot y*n for two distinct "regions" and then overlay the levels of a second factor "m" onto the plots for each region. For the example below, I will have 1 column and 2 rows with each row corresponding to a "region". I then overlay the values of y for the second factor "m" onto each plot for the respective "region". A legend is created at the top of the graph listing the linetype (lty) for each level of "m".
THE PROBLEM: When I copy the graph in s-plus and paste it directly into Microsoft word, the legend identifying the line types become almost transparent, so when I print it out you can hardly identify which line type corresponds to the level of "m". The lty's in the graph look great, so no problem there. I have also tried exporting the graph as a .jpeg, bmp and the same thing occurs.
THE SOLUTION: How can I make the lty's in the legend more identifiable? I cannot increase the line width (lwd) b/c the values between the levels of "m" are so small that you could not distinguish between the levels.
This is not the case for the example below b/c I only have 2 levels of "m", but for the entire data set I have 6 levels and it just looks like a great big mess if I increase the LWD.
Any suggestions as to how I can remedy this problem? Is there perhaps an easier way to get the graph that I am interested in?
Thanks for the help...
Lewis Jordan
"Plots"<-
structure(list(n=c(20, 40, 60, 80, 100, 20, 40, 60, 80, 100,
20, 40, 60, 80, 100, 20, 40, 60, 80, 100),
region = c("North", "North", "North", "North", "North", "North", "North", "North", "North", "North",
"South","South", "South", "South", "South", "South", "South", "South", "South", "South"),
m=c( 5, 5, 5, 5, 5, 30, 30, 30, 30, 30,
&n
bsp; 5, 5, 5, 5, 5, 30, 30, 30, 30, 30),
y=c( .005, .004, .003, .003, .002, .004, .003, .002, .002, .002,
.006, .004, .003, .003, .003, .005, .003, .003, .002, .002)),
.Names=c("n","Region","m","y"),
row.names=as.character(1:20),
class="data.frame")
Plots[1:3,]
Plots$groupM<-ordered(Plots$m,levels=c(5,30),c("5","30"));
gt$lty<-c(1,3)
xlim <- range(Plots$n)
ylim <- range(Plots$y)
trellis.device(color=F)
xyplot(y~n|Region,data="" groups=groupM,xlim=xlim,ylim=ylim,subscripts=T,subset=T,layout=c(1,2),
scales=list(relation="free"),
panel=function(x,y,subscripts,groups)
{
panel.grid()
panel.superpose(x,y,subscripts,groups,type="l",lty=c(1,3))
},
main="",
xlab="n (Stands)",
ylab="Standard Error",
key=list(lines=gt,
text=list(c("5","30")),
columns=2,),
strip=function(...) strip.default(...,style=1)
) |