[R] Need help with xspline error message in R 2.6.1 on Windows XP

Erik Iverson iverson at biostat.wisc.edu
Mon Jan 14 21:37:09 CET 2008


Walter,

Try adding the argument default.units = "native" to your grid.xspline call.


Erik Iverson


Anderson, Walter wrote:
> I am attempting to generate a series of plots with cubic B-Splines and
> the code I am using is generating the following error message that I
> have been unable to resolve.
> 
> Error in grid.Call.graphics("L_xspline", x$x, x$y, x$shape, x$open,
> x$arrow,  :   add_point - reached MAXNUMPTS (25200)
> 
> This is the code that generates this error.
> 
> plot.results <- function(dfn) {
> 	dt <- read.table(dfn)
> 	shp <- rep(1, as.numeric(length(dt$year)))
> 	shp[1] <- shp[as.numeric(length(dt$year))] <- 0
> 	grid.rect()
> 	pushViewport(plotViewport(c(5.1, 4.1, 4.1, 2.1)))
> 	pushViewport(dataViewport(dt$year, dt$mean))
> 	grid.rect()
> 	grid.xaxis()
> 	grid.yaxis()
> 	grid.points(dt$year, dt$mean, pch=16)
> 	grid.xspline(dt$year, dt$mean, shape=shp, open=TRUE,
> gp=gpar(col="red"))
> 	popViewport(2)
> }
> 
> plot.results("section01_model_results.dat")
> 
> And here is the data that is being used:
> 
>   year  mean
> 1 2010  6738
> 2 2015  8793
> 3 2020 11699
> 4 2025 15951
> 5 2030 22492
> 6 2035 33289
> 7 2040 52917
> 
> I have tried changing the values in the shape vector and have noticed
> that a value of 0 does not generate the error; however, it also does not
> plot a line or curve.  I have also tried values for the shape vector of
> -1, 0.1, 0.2, and 0.5.  
> 
> I think my problem may be unit related, but am unclear how to specify
> that my control points are user units (or to convert user units to what
> ever xspline might prefer).
> 
> Walter Anderson
> 
> ______________________________________________
> R-help at r-project.org 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.




More information about the R-help mailing list