1

I have a matrix of 366*1055 dimension. I wanted to subtract each value (each row consists of 1055 elements) in a row with its respective rowMedian value. How to go about it ??

satty
  • 23
  • 1
  • 1
  • 5

2 Answers2

2

Using the base package. The matrix m1 is from akrun's answer.

Explanation

We can use the apply function to apply a function, median in this case, to every row in the matrix m1. We set the second argument to 1 to indicate that the function will be applied over rows (row-by-row). The apply functions are very useful in R. For more details, type ?apply in the console or check this thorough answer by Joran.

m1 - apply(m1, 1, median)

Output 

      [,1] [,2] [,3] [,4] [,5]
 [1,]   -1    3   -3    0    1
 [2,]   -1    0    2   -1    0
 [3,]    1    0    1   -3   -4
 [4,]    3    4   -1   -1    0
 [5,]    3    0   -1    6   -3
 [6,]    3    3    0   -4   -6
 [7,]    0   -1   -2    5    1
 [8,]    2   -5    0    1   -2
 [9,]    2   -1    0    0   -4
[10,]    0   -1   -2    7    1
Community
  • 1
  • 1
mpalanco
  • 12,960
  • 2
  • 59
  • 67
  • While this answer is correct, it would be worth adding some explanatory text so that later readers with less knowledge can understand WHY this is correct and how it works. – kdopen Oct 15 '15 at 19:18
  • @kdopen You're right. I've just edited my answer to provide an explanation and a couple of references. Thank you. – mpalanco Oct 15 '15 at 20:40
1

We can use rowMedians from matrixStats to calculate the rowwise median and subtract it from the matrix 'm1'.

library(matrixStats)
m1 - rowMedians(m1, na.rm=TRUE)

data

set.seed(24)
m1 <- matrix(sample(0:9, 10*5, replace=TRUE), ncol=5)
akrun
  • 874,273
  • 37
  • 540
  • 662