using anonymous functions in R with multiple arguments

Question

I'm attempting to generate new variables in a data frame that are conditional on two (or more) other variables in the data frame. I believe that the looping functions in R (i.e., lapply, sapply, etc.) are useful and efficient for this purpose. Something is not right, however, with my approach, and I can't figure out what.

M <- data.frame(x=c("A", "A", "B", "B"), y=c(1,2,1,2))

Using this data frame, I would like to generate a new column z, containing logicals that are TRUE iff both x == "A" and y == 1. The following code is the best I can come up with here, but only seems to evaluate my first condition.

M$z <- sapply(M$x, function(x,y) if((x == "A") && (y == 1)) T else F, M$y)

• Can this code be fixed for my purpose?
• Is there a better way of doing this in R, perhaps using other looping functions?

Answer 1

This is a task for transform function

transform(M, z=ifelse((x == "A") & (y == 1), T, F))
  x y     z
1 A 1  TRUE
2 A 2 FALSE
3 B 1 FALSE
4 B 2 FALSE

I think an even simpler approach would be

M$z <- with(M, (x == "A") & (y == 1))
M
  x y     z
1 A 1  TRUE
2 A 2 FALSE
3 B 1 FALSE
4 B 2 FALSE

Answer 2

Take a look at mapply:

> M$z <- mapply(M$x,M$y, FUN=function(x,y) if((x == "A") && (y == 1)) T else F)
> M
  x y     z
1 A 1  TRUE
2 A 2 FALSE
3 B 1 FALSE
4 B 2 FALSE

Apropos, this has nothing to do with anonymous functions and everything to do with applying with multiple arguments. If you named the function it would still not work in any of the single-argument apply variants.

The other way to do this would be to ddply by row, or split your data.frame into a list with each row being a separate entry.