How to use functionals with a list or vector of functions?-CodePudding

I want to use functionals to calculate different summary statistics onto an data set / simulation.

choose_summary <- function(f) f(1:100)
choose_summary(mean)
choose_summary(median)
choose_summary(max)

Thats an easy example on how this should work. I want to use a list of functions to be applied to the data, but it doesnt work the way I expected it to.

sum_fun_list <- list(mean, max, median)
choose_summary(sum_fun_list [1])

Output: Error in f(1:100) : could not find function "f"

What I tried to to is get a element of the list and use it as input for my function, but that doesnt seem to work. It looks like R doesnt recognize the input as a function. I have no idea how to work around that. I tried different approaches like unlist() or as.function() but nothing seems to do the trick. Any idea how to make this work?

Thanks!

CodePudding user response：

do.call is your friend here.

We can make a list of functions and a vector:

funs <- list(mean,median,max)
my_vec <- list(1:100)

Then use sapply to call any function in funs with the argument my_vec

sapply(funs,do.call,my_vec)

CodePudding user response：

The problem is in the way you're indexing the list of functions. This works:

choose_summary(sum_fun_list[[1]])
[1] 50.5

and so does this:

sapply(sum_fun_list, choose_summary)
[1]  50.5 100.0  50.5

When you index a list using [...], as in sum_fun_list[1] you return a list with one element. When you index a list with [[...]] you get the contents of that element (which is the function in your case).