How to find the row median of columns having same name in data.table object in R?

To find the row median of columns having same name in data.table object in R, we can follow the below steps −

  • First of all, create a data.table with some columns having same name.

  • Then, use tapply along with colnames and median function to find the row median of columns having same name.

Example

Create the data.table

Let’s create a data.table as shown below −

library(data.table)
DT<-
data.table(x=rpois(25,5),y=rpois(25,10),x=rpois(25,5),y=rpois(25,2),x=rpois(25,4),y=rpo is(25,10),check.names=FALSE)
DT

Output

On executing, the above script generates the below output(this output will vary on your system due to randomization) −

     x  y  x y x  y
1:   4  15 6 1 6  9
2:   4  11 3 3 4  9
3:   1  6  8 2 2 12
4:   6  9  4 2 6 12
5:   8 10  1 1 2  9
6:   6  9  7 4 2  4
7:   8  8  3 2 1  8
8:   2 10  7 2 5 16
9:   7  5  8 1 5 9
10:  5 11  9 5 7 10
11:  3  4  1 5 6  6
12: 10 10  6 3 6 13
13:  7  5  4 2 3  8
14:  4  9  0 2 5 14
15:  6 11  8 2 4 10
16:  6  8  3 3 7  7
17:  7  8  4 2 1  8
18:  8  9  7 4 3 15
19:  4 16  5 1 3  7
20:  7 15  4 1 3 12
21:  6 10  5 1 1  9
22:  4 11  3 0 2  9
23:  4 12  4 3 4 17
24:  4  8  2 1 5  6
25:  5  9  1 0 2  5
     x  y  x y x  y

Find the row median of columns having same name

Using tapply along with colnames and median function to find the row median of columns having same name in data.table object DT −

library(data.table)
DT<-
data.table(x=rpois(25,5),y=rpois(25,10),x=rpois(25,5),y=rpois(25,2),x=rpois(25,4),y=rpo is(25,10),check.names=FALSE)
t(apply(DT,1, function(x) tapply(x,colnames(DT),median)))

Output

      x y
[1,]  6 9
[2,]  4 9
[3,]  2 6
[4,]  6 9
[5,]  2 9
[6,]  6 4
[7,]  3 8
[8,]  5 10
[9,]  7 5
[10,] 7 10
[11,] 3 5
[12,] 6 10
[13,] 4 5
[14,] 4 9
[15,] 6 10
[16,] 6 7
[17,] 4 8
[18,] 7 9
[19,] 4 7
[20,] 4 12
[21,] 5 9
[22,] 3 9
[23,] 4 12
[24,] 4 6
[25,] 2 5

Updated on: 10-Nov-2021

43 Views

Kickstart Your Career

Get certified by completing the course

Get Started
Advertisements