[R] maximum value replacement

Bert Gunter gunter.berton at gene.com
Tue Oct 29 16:30:13 CET 2013

```To Eliza: What if the max in a column is not unique?

Given the small size of A, the solution given by Arun seems completely
adequate. However, I was wondering if it could be done without the
R-level loop in sapply by taking advantage of pmax() . Of course it
can. Here's code to illustrate how:

A<- matrix(sample(15),nr=3)
A
B <- matrix(0,nr=1,nc=5)
m <- do.call(pmax,data.frame(t(A)))
d <- dim(A)
A[which(A==matrix(m,nr=d[1],nc=d[2],byrow=TRUE))] <- B
A

Note that:

1. This does not generalize to functions other than max or min, afaik.
2. I don't even know if it would be faster for large data, because the
data.frame call may slow things down.

But it is fully vectorized (I think). So for illustration only, maybe...

Cheers,
Bert

On Tue, Oct 29, 2013 at 7:27 AM, arun <smartpink111 at yahoo.com> wrote:
> Hi,
> Try:
>  sapply(seq_len(ncol(A)),function(i) {indx <- which(A[,i]%in% max(A[,i])); A[,i][indx] <- B[,i]; A[,i]})
>
> A.K.
>
>
>
>
> On Tuesday, October 29, 2013 10:16 AM, eliza botto <eliza_botto at hotmail.com> wrote:
> Dear Users,
> I have two matrices, one with 12 rows and 124 columns(A) and the other with 1 row and 124 column(B). i want to replace the maximum value in all columns of A with each (single) column value of B.
> How can i do it??
>
> Eliza
>     [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> and provide commented, minimal, self-contained, reproducible code.
>
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> and provide commented, minimal, self-contained, reproducible code.

--

Bert Gunter
Genentech Nonclinical Biostatistics

(650) 467-7374

```