Dear Chad, your problem is linked to (1) the function returning NaNs from x values greater than 50, and (2) the fact that the function is estimated on a predefined number of points.
Calling plot for a function object is basically a wrapper for curve(). Your function g() is evaluated on the whole xlim domain, which will return NaN values for x>50 (Try g(60) ). In addition, curve() splits the x interval (here from 0 to 60) into a predifined number of points (n=101 is the default, see help(curve)) at which the function is estimated. In your code, the function is estimated at values x <- seq(0, 60, length=101), and g(x) that are not NaN are plotted. The largest x value (from the sequence) that doesn't return a NaN is max(x[!is.nan(g(x))]), which is 49.8. One way to solve it is to explicitly specify the domain used to estimate the function, by using the from and to arguments that are passed to curve(): #Figure 2, with xlim beyond the radius of the circle plot(g,axes=F,from=0, to =50, xlim=c(0, 60), ylim=c(0,60)) axis(1,pos=0) axis(2,pos=0) HTH Matthieu Matthieu Dubois Post-doctoral researcher Psychology Department Université Libre de Bruxelles ______________________________________________ [email protected] 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.
