Finding the maximum value for each row among 3 columns in R
Use data.table :)
library(data.table)
x = c(1,2,3,4,5 )
y = c(2,3,3,1,1 )
z = c(4,3,2,1,1 )
dt<-data.table(x,y,z)
dt[, max:=pmax(x,y,z)]
dt
You can use the apply
function for this like so:
df$max<-apply(X=df, MARGIN=1, FUN=max)
The MARGIN=1
argument indicated that for every row in X
you wish to apply the function in FUN
. If you use MARGIN=2
it will be by column or MARGIN=c(1,2)
it will be both rows and columns.
Try:
df$max <- do.call(`pmax`, df)
df
# x y z max
#1 1 2 4 4
#2 2 3 3 3
#3 3 3 2 3
#4 4 1 1 4
#5 5 1 1 5
Benchmarks
set.seed(49)
df <- as.data.frame(matrix(sample(0:20, 1e5*3,replace=TRUE), ncol=3))
f1 <- function() df$max <- apply(df, 1, max)
f2 <- function() df$max <- do.call(`pmax`, df)
f3 <- function() setDT(df)[, max:=pmax(V1,V2,V3)]
library(microbenchmark)
microbenchmark(f1(), f2(),f3(), unit="relative", times=25)
#Unit: relative
# expr min lq median uq max neval
# f1() 48.143635 48.287875 46.031638 32.868138 8.922203 25
# f2() 1.269581 1.373479 1.654625 2.324896 1.182107 25
# f3() 1.000000 1.000000 1.000000 1.000000 1.000000 25