apply over matrix by column - any way to get column name?

Question

I'm calling apply(data, 2, function(x) {...}) where data is a matrix. Is there any way for the function to know the name of the column whose values are present in x? Or perhaps there's another function to accomplish the same task?

score 30 · Accepted Answer · answered Oct 06 '11 at 22:10

30

One common way to deal with this is to loop over the column names themselves:

m <- matrix(runif(16),4,4)
colnames(m) <- letters[1:4]
sapply(colnames(m),function(x){mean(m[,x])})
        a         b         c         d 
0.4720319 0.4992337 0.5026318 0.3514267

answered Oct 06 '11 at 22:10

joran

169,992
32
429
468

5

Bah, beat me by 5 seconds! When will I learn... answer first, then edit. ;-) – Joshua Ulrich Oct 06 '11 at 22:11
Would `sapply(colnames(m),function(x, data){mean(data[,x])}, m)` be slightly better as it explicitly passes data into the function? – Joseph Crispell Mar 02 '21 at 15:36

score 18 · Answer 2 · answered Oct 06 '11 at 22:10

You can use sapply on colnames(data):

set.seed(21)
Data <- matrix(rnorm(10),5,2)
colnames(Data) <- c("one","two")
sapply(colnames(Data), function(x) sum(Data[,x]))
#       one       two 
#  3.987540 -2.010875 
colSums(Data)
#       one       two 
#  3.987540 -2.010875

apply over matrix by column - any way to get column name?

2 Answers2

Linked

Related