Do subtraction between columns of a data.frame

Admin 1
Viewed 60

The database below presents the ranking of three methods, which I called M1, M2 and M3. Therefore, I would like to make a new dataset with the difference between the ranks of two methods. In this case, the following relation would be, M1 and M2, M1 and M3, M2 and M3. Therefore, the dataset will have three columns (n, M1-M2, M1-M3 and M2-M3).

result<-structure(list(n = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 
12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 
28, 29), M1 = c(1L, 29L, 28L, 27L, 25L, 26L, 24L, 20L, 21L, 
22L, 23L, 15L, 12L, 17L, 18L, 19L, 16L, 13L, 14L, 5L, 6L, 7L, 
8L, 9L, 10L, 11L, 4L, 2L, 3L), M2 = c(1, 29, 28, 27, 26, 25, 
24, 23, 22, 21, 20, 15, 12, 19, 18, 17, 16, 14, 13, 11, 10, 9, 
8, 7, 6, 5, 4, 3, 2), M3 = c(1L, 29L, 28L, 27L, 25L, 26L, 24L, 
20L, 21L, 22L, 23L, 15L, 12L, 17L, 18L, 19L, 16L, 13L, 14L, 5L, 
6L, 7L, 8L, 9L, 10L, 11L, 4L, 2L, 3L)), class = "data.frame", row.names = c(NA,-29L))
  
    > result
    n M1 M2 M3
1   1  1  1  1
2   2 29 29 29
3   3 28 28 28
4   4 27 27 27
5   5 25 26 25
6   6 26 25 26
7   7 24 24 24
8   8 20 23 20
9   9 21 22 21
10 10 22 21 22
11 11 23 20 23
12 12 15 15 15
13 13 12 12 12
14 14 17 19 17
15 15 18 18 18
16 16 19 17 19
17 17 16 16 16
18 18 13 14 13
19 19 14 13 14
20 20  5 11  5
21 21  6 10  6
22 22  7  9  7
23 23  8  8  8
24 24  9  7  9
25 25 10  6 10
26 26 11  5 11
27 27  4  4  4
28 28  2  3  2
29 29  3  2  3
r
4 Answers

We can try

cbind(
  result,
  `colnames<-`(
    combn(result[-1], 2, function(v) v[[1]] - v[[2]]),
    combn(names(result)[-1], 2, function(v) paste0(v, collapse = "-"))
  )
)

which gives

    n M1 M2 M3 M1-M2 M1-M3 M2-M3
1   1  1  1  1     0     0     0
2   2 29 29 29     0     0     0
3   3 28 28 28     0     0     0
4   4 27 27 27     0     0     0
5   5 25 26 25    -1     0     1
6   6 26 25 26     1     0    -1
7   7 24 24 24     0     0     0
8   8 20 23 20    -3     0     3
9   9 21 22 21    -1     0     1
10 10 22 21 22     1     0    -1
11 11 23 20 23     3     0    -3
12 12 15 15 15     0     0     0
13 13 12 12 12     0     0     0
14 14 17 19 17    -2     0     2
15 15 18 18 18     0     0     0
16 16 19 17 19     2     0    -2
17 17 16 16 16     0     0     0
18 18 13 14 13    -1     0     1
19 19 14 13 14     1     0    -1
20 20  5 11  5    -6     0     6
21 21  6 10  6    -4     0     4
22 22  7  9  7    -2     0     2
23 23  8  8  8     0     0     0
24 24  9  7  9     2     0    -2
25 25 10  6 10     4     0    -4
26 26 11  5 11     6     0    -6
27 27  4  4  4     0     0     0
28 28  2  3  2    -1     0     1
29 29  3  2  3     1     0    -1
user_121587570 
result = cbind(n=result$n,(result[,c(2:4)] - result[,c(3,4,2)])[,c(1,3,2)])
names(result)=c("n", "M1-M2", "M1-M3", "M2-M3")

Output:

    n M1-M2 M1-M3 M2-M3
1   1     0     0     0
2   2     0     0     0
3   3     0     0     0
4   4     0     0     0
5   5    -1     0     1
6   6     1     0    -1
7   7     0     0     0
8   8    -3     0     3
9   9    -1     0     1
10 10     1     0    -1
11 11     3     0    -3
12 12     0     0     0
13 13     0     0     0
14 14    -2     0     2
15 15     0     0     0
16 16     2     0    -2
17 17     0     0     0
18 18    -1     0     1
19 19     1     0    -1
20 20    -6     0     6
21 21    -4     0     4
22 22    -2     0     2
23 23     0     0     0
24 24     2     0    -2
25 25     4     0    -4
26 26     6     0    -6
27 27     0     0     0
28 28    -1     0     1
29 29     1     0    -1
user_44475400

For a tidyverse solution you can use mutate() from the dplyr package to create new columns in your data.frame:

library(dplyr)

newDF <- result %>% 
  mutate(D1 = M1-M2, D2 = M1-M3, D3 = M2 - M3)

Output:

    n M1 M2 M3 D1 D2 D3
1   1  1  1  1  0  0  0
2   2 29 29 29  0  0  0
3   3 28 28 28  0  0  0
4   4 27 27 27  0  0  0
5   5 25 26 25 -1  0  1
6   6 26 25 26  1  0 -1
7   7 24 24 24  0  0  0
8   8 20 23 20 -3  0  3
9   9 21 22 21 -1  0  1
10 10 22 21 22  1  0 -1
11 11 23 20 23  3  0 -3
12 12 15 15 15  0  0  0
13 13 12 12 12  0  0  0
14 14 17 19 17 -2  0  2
15 15 18 18 18  0  0  0
16 16 19 17 19  2  0 -2
17 17 16 16 16  0  0  0
18 18 13 14 13 -1  0  1
19 19 14 13 14  1  0 -1
20 20  5 11  5 -6  0  6
21 21  6 10  6 -4  0  4
22 22  7  9  7 -2  0  2
23 23  8  8  8  0  0  0
24 24  9  7  9  2  0 -2
25 25 10  6 10  4  0 -4
26 26 11  5 11  6  0 -6
27 27  4  4  4  0  0  0
28 28  2  3  2 -1  0  1
29 29  3  2  3  1  0 -1
user_7686190

Its a dataframe so you should be able to do something like

result$M1_M2 <- result$M1 - result$M2

or something similar

user_183771230
Related