[R] function names in a vector used by for (){} character problem ?

Duncan Murdoch murdoch at stats.uwo.ca
Thu Jul 20 21:56:54 CEST 2006


On 7/20/2006 3:42 PM, bunny , lautloscrew.com wrote:
> Hi there,
> 
> i´m have vector of kernels. just like:
> kernels = c('gauss','epan','rectangular')
> 
> i know there are density.default$kernels, but thats not my question  
> here. my own kernel functions are running and working.
> my problem is the following is not working:
> 
> 
> dev.off()
> par(mfrow=c(3,3))
> 
> 
> for(i in 1:length(bw))
> {
>          for(j in 1:length(kernels))
>          {
>          par(mfg = c(i, j))	       	
> 		nawaline2(eruptions,waiting,kernels[j],bw[i],1000)
> 		# FYI: the following worked but ofcourse all rows were the same
> 		# nawaline2(eruptions,waiting,gauss,bw[i],1000)
> 
>          }
> }
> 
> # here are the standard arguments for nawaline, kern exspects a  
> function like gaussian, epan or rectangular.
> nawaline2(xi,yi,kern,h,N=1000)
> 
> there´s bw vector also that works well. the problem seems  that my  
> kernels vector ist a vector of characters.
> I want to loop it using for ;  but every time it runs through, there 
> ´s an error, that kern can´t be found.
> i think it´s only beacause  i don´t know how to get gauss instead of  
> "gauss" and so on...

You could set up kernels as

kernels <- list(gauss, epan, rectangular)

and then pass kernels[[j]] instead of kernels[j].

Or your nawaline2 function could have something like

if (is.character(kern)) kern <- get(kern, envir=parent.frame())

to go looking for the object with the name stored in kern.

Duncan Murdoch

> 
> has anybody an idea ? thanks so much in advance !!
> 
> 
> 
> 
> 	[[alternative HTML version deleted]]
> 
> 
> 
> ------------------------------------------------------------------------
> 
> ______________________________________________
> R-help at stat.math.ethz.ch 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