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
CodePudding user response:
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
CodePudding user response:
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
CodePudding user response:
result = cbind(n=result$n,(result[,c(2:4)] - result[,c(3,4,2)])[,c(1,3,2)])
names(result)=c("n", "M1-M2", "M1-M3", "M2-M3")
Output:
n M1-M2 M1-M3 M2-M3
1 1 0 0 0
2 2 0 0 0
3 3 0 0 0
4 4 0 0 0
5 5 -1 0 1
6 6 1 0 -1
7 7 0 0 0
8 8 -3 0 3
9 9 -1 0 1
10 10 1 0 -1
11 11 3 0 -3
12 12 0 0 0
13 13 0 0 0
14 14 -2 0 2
15 15 0 0 0
16 16 2 0 -2
17 17 0 0 0
18 18 -1 0 1
19 19 1 0 -1
20 20 -6 0 6
21 21 -4 0 4
22 22 -2 0 2
23 23 0 0 0
24 24 2 0 -2
25 25 4 0 -4
26 26 6 0 -6
27 27 0 0 0
28 28 -1 0 1
29 29 1 0 -1
CodePudding user response:
Its a dataframe so you should be able to do something like result$M1_M2 <- result$M1 - result$M2 or something similar