using the dots argument in replicate expression fails - why exactly?

Question

I want to use the dots arguments (...) in an expression in replicate. I noticed that they do not have an effect doing it this way:

foo <- function(k=1, n=10, ...) 
{
  replicate(k, rnorm(n, ...))
}

foo(2, mean=100)

Results do not have a mean of 100.

           [,1]       [,2]
[1,]  0.2859647 -0.1046510
[2,] -0.7867414  0.5347617

Using a wrapper, however, will work.

foo2 <- function(k=1, n=10, ...) 
{
  f <- function() rnorm(n, ...)
  replicate(k, f())
}

foo2(2, mean=100)

Now, the results do have a mean of 100.

         [,1]     [,2]
[1,] 100.9644 100.6287
[2,] 100.0804 101.0218

What exactly are the mechanics going on here during the evaluation of the expr argument in replicate. Can anyone give an explanation of the behaviour?

The joys of non-standard evaluation. Just use `sapply` directly in this specific example. — Roland, Mar 24 '16 at 09:28

score 1 · Answer 1 · answered Mar 24 '16 at 09:24

This behavior is documented:

If expr is a function call, be aware of assumptions about where it is evaluated, and in particular what ... might refer to. You can pass additional named arguments to a function call as additional named arguments to replicate: see ‘Examples’.

And in the "Examples" section:

## use of replicate() with parameters:
foo <- function(x = 1, y = 2) c(x, y)
# does not work: bar <- function(n, ...) replicate(n, foo(...))
bar <- function(n, x) replicate(n, foo(x = x))
bar(5, x = 3)

using the dots argument in replicate expression fails - why exactly?

1 Answers1