R help - Nov 2009 - Using processed objects as arguments of a function

Dear R users,

I wish to utilise processed and saved objects as arguments of a function.

Specifically, I have created objects using *"assign"* &
*"paste"* functions
with an incremental index i, the names of the objects are:

    fund1, fund2, fund3,....., fund80,..... (where the numerical value
increments according to the index i & class of these objects are dataframes)

I wish to collapse these objects row wisely using *"rbind"* function.

paste("fund", 1:i, sep = "") results in list of objects as
characters...  &
get(paste("fund", 1:i, sep = "")) outputs fund1...

Are there any methods to use these objects as an argument of "rbind"
to
collapse the dataframes?

Your expertise in resolving this issue would be highly appreciated.





Steven

	[[alternative HTML version deleted]]

On 11/02/2009 05:04 PM, Steven Kang wrote:
> Dear R users,
>
> I wish to utilise processed and saved objects as arguments of a function.
>
> Specifically, I have created objects using *"assign"*& 
*"paste"* functions
> with an incremental index i, the names of the objects are:
>
>      fund1, fund2, fund3,....., fund80,..... (where the numerical value
> increments according to the index i&  class of these objects are
dataframes)
>
> I wish to collapse these objects row wisely using *"rbind"*
function.
>
> paste("fund", 1:i, sep = "") results in list of objects
as characters...&
> get(paste("fund", 1:i, sep = "")) outputs fund1...
>
> Are there any methods to use these objects as an argument of
"rbind" to
> collapse the dataframes?
>
> Your expertise in resolving this issue would be highly appreciated.
Hi Steven,
There is probably a neater way to construct the list of dataframes, but 
this will probably do what you want:

dnames<-paste("fund",1:nfunds,sep="")
makelist<-function(x) {
  nitems<-length(x)
  newlist<-vector("list",nitems)
  for(item in 1:nitems) newlist[[item]]<-get(x[item])
  return(newlist)
}
dflist<-makelist(dfnames)
do.call("rbind",dflist)

Of course all of the dataframes must have the same number of columns or 
the result will be messy or not there at all.

Jim