pairwise comparison with all vectors of a list-CodePudding

I have, for example, a list of 5 vectors and would like to compare each one of them. Example

L = list(c(1:5), c(-1:-5), c(3:7), c(-4:-8), c(5:9))
[[1]]
[1] 1 2 3 4 5

[[2]]
[1] -1 -2 -3 -4 -5

[[3]]
[1] 3 4 5 6 7

[[4]]
[1] -4 -5 -6 -7 -8

[[5]]
[1] 5 6 7 8 9

I have also a function to make pairwise comparisons between the 5 elements of the list:

foo = function(x, y) t(x) %*% abs(y)

I would like to apply foo to make pairwise comparisons of my list. For example: Pairwise comparisons (the 'x' and 'y' of foo):

[[1]]and [[2]]
[[1]]and [[3]]
[[1]]and [[4]]
[[1]]and [[5]] 
[[2]]and [[3]]
[[2]]and [[4]]
[[2]]and [[5]]
[[3]]and [[4]]
[[3]]and [[5]]
[[4]]and [[5]]
...
My original file has 1000 elements

I tried using: lapply(L, foo) but I have the following message error: ```argument "y" is missing, with no default````

How can I apply foo to my list?

CodePudding user response：

Perhaps a combination of combn, apply and lapply?

L = list(c(1:5), c(-1:-5), c(3:7), c(-4:-8), c(5:9))
foo = function(x, y) t(x) %*% abs(y)

setNames(lapply(apply(combn(seq_along(L), 2), 2, function(x) L[x]), 
       function(z) foo(z[[1]], z[[2]])),
       apply(combn(seq_along(L), 2), 2, paste, collapse = "-"))
#> $`1-2`
#>      [,1]
#> [1,]   55
#> 
#> $`1-3`
#>      [,1]
#> [1,]   85
#> 
#> $`1-4`
#>      [,1]
#> [1,]  100
#> 
#> $`1-5`
#>      [,1]
#> [1,]  115
#> 
#> $`2-3`
#>      [,1]
#> [1,]  -85
#> 
#> $`2-4`
#>      [,1]
#> [1,] -100
#> 
#> $`2-5`
#>      [,1]
#> [1,] -115
#> 
#> $`3-4`
#>      [,1]
#> [1,]  160
#> 
#> $`3-5`
#>      [,1]
#> [1,]  185
#> 
#> $`4-5`
#>      [,1]
#> [1,] -220

^{Created on 2021-10-25 by the reprex package (v2.0.0)}

CodePudding user response：

What about outer?

> outer(L, L, Vectorize(foo))
     [,1] [,2] [,3] [,4] [,5]
[1,]   55   55   85  100  115
[2,]  -55  -55  -85 -100 -115
[3,]   85   85  135  160  185
[4,] -100 -100 -160 -190 -220
[5,]  115  115  185  220  255