I guess it is clipped in this case. Use par("usr") after ts.plot() to see what the user coordinates are, and whether (1, -0.3) is a sensible location to place the legend.
Uwe Ligges
Just to add to Uwe's good suggestion -- what I usually do is
opar <- par(no.readonly=TRUE) par(usr=c(0,1,0,1))
Then pick legend co-ordinates, knowing that x and y on the plot canvas now range from 0 to 1. For somewhere near the top-right corner...
legend(0.8,0.8, ...) par(opar) ## when done with the plot, to restore the co-ords.
Another choice is instead of the "par" calls...
legend(locator(1), ...)
and click on the plot where you want the legend.
Thirdly, the Hmisc package has a function largest.empty(), which can find the "best" place to put a legend in a plot automagically.
Cheers
Jason
______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
