R help - Jan 2009 - How can pairs of values be stored in a 2D matrix ?

My question is motivated by dealing with pairs of values, like (3,12.5), 
(16,2.98), and so on
that are mapped to a cartesian plain (Time, Frequence) 
I miss C multidimensional arrays. I am trying to simulate the 3rd dimension by
declaring a matrix of lists.
R seems to accept the definition of such a matrix. But I'm experiencing some
difficulty at storing pairs of data
in it. In the following are some of my attempts.  
I would greatly appreciate some help. 
> A <- matrix(data ="list", nrow=2, ncol=4)
> A
     [,1]   [,2]   [,3]   [,4]  
[1,] "list" "list" "list" "list"
[2,] "list" "list" "list"
"list"
> A[[1]] <- list(data="12.3",variable="d1")
Error in A[[1]] <- list(data = "12.3", variable = "d1") :
  more elements supplied than there are to replace
> A[1,2] <- list(data="12.3",variable="d1")
Error in A[1, 2] <- list(data = "12.3", variable = "d1")
:
  number of items to replace is not a multiple of replacement
length
> A[[1,2]] <- list(data="12.3",variable="d1")
Error in A[[1, 2]] <- list(data = "12.3", variable =
"d1") :
  more elements supplied than there are to replace
> A[[1,1]] <- list(data="12.3",variable="d1")
Error in A[[1, 1]] <- list(data = "12.3", variable =
"d1") :
  more elements supplied than there are to replace
> A[[1]][1] <- 10
> A[[1]][2] <- "f1-t1"
Error in A[[1]][2] <- "f1-t1" : 
  more elements supplied than there are to replace
> A[[1]][2] <- "d1"
Error in A[[1]][2] <- "d1" : 
  more elements supplied than there are to replace

Regards,
Maura










tutti i telefonini TIM!


	[[alternative HTML version deleted]]

On Mon, Jan 5, 2009 at 6:30 AM,  <mauede at alice.it>
wrote:
> My question is motivated by dealing with pairs of values, like (3,12.5), 
(16,2.98), and so on
> that are mapped to a cartesian plain (Time, Frequence)
> I miss C multidimensional arrays. I am trying to simulate the 3rd dimension
by declaring a matrix of lists.
I think the standard R way to do this would be with a 3-dimensional
array.  Many functions are designed to work well with n-dimensional
arrays, including apply and the plyr package.

But R does allow you to make a matrix of lists.
>> A <- matrix(data ="list", nrow=2, ncol=4)
This creates a matrix of strings (character vector).  What you want is
a matrix of lists, something like this:

  A<-matrix(data=list(), nrow=2,ncol=4)

Then I think all your examples below work.

           -s

PS Note that it does *not* work to write data=NULL, because NULL is
not of class/mode 'list', even though the resulting matrix is filled
with NULL.