CODEWITHSUNDEEP
r
flee
flee

Reputation: 1335

Apply function to columns in a list of data frames and append results

I want to apply a function to a list of data frames. the function takes elements from two columns in each data frame adds them and then adds the output to each data frame in a new column.

Create dummy data:

df.1 <- data.frame(data=cbind(rnorm(5, 0), rnorm(5, 2), rnorm(5, 5)))
df.2 <- data.frame(data=cbind(rnorm(5, 0), rnorm(5, 2), rnorm(5, 5)))

names(df.1) <- c("a", "b", "c")
names(df.2) <- c("a", "b", "c")

ls.1<- list(df.1,df.2)
names(ls.1) <- c("cat", "dog")
ls.1

have a look at the data:

> ls.1
$cat
          a        b        c
1 0.7031868 1.730499 4.286386
2 0.1527551 2.794084 4.348707
3 1.1151157 0.154562 4.647605
4 0.5786497 1.407386 4.118078
5 0.9223104 2.995469 5.065981

$dog
            a         b        c
1  0.04024872 1.6760609 5.013490
2  0.18095899 2.1015250 3.452313
3 -0.86588484 2.1371948 6.389203
4 -0.39499567 0.5996709 5.399724
5 -1.31850123 3.0058084 5.530989

Pseudo code of what I want to do:

 my.fun <- function(b, c) {
  out.put <- b + c
  ls.1[i]$d <- out.put
}

What I want the output to look like:

> ls.1
$cat
          a        b        c        d
1 0.7031868 1.730499 4.286386 6.689551
2 0.1527551 2.794084 4.348707 5.553838
3 1.1151157 0.154562 4.647605 8.526398
4 0.5786497 1.407386 4.118078 5.999395
5 0.9223104 2.995469 5.065981 8.536797

$dog
            a         b        c        d
1  0.04024872 1.6760609 5.013490 6.689551
2  0.18095899 2.1015250 3.452313 5.553838
3 -0.86588484 2.1371948 6.389203 8.526398
4 -0.39499567 0.5996709 5.399724 5.999395
5 -1.31850123 3.0058084 5.530989 8.536797

I think it should be easily achievable with mapply or something similar but I I cannot figure it out....

Upvotes: 4

Views: 9882

Answers (3)

S. Elzwawi
S. Elzwawi

Reputation: 541

Here is an alternative solution using a for loop

for (e in  names(ls.1)) {
   ls.1[[e]]$d <- ls.1[[e]]$b + ls.1[[e]]$c
}

$cat
           a           b        c        d
1  1.2720987  2.03251108 4.395936 6.428447
2  0.1048982  1.47652374 3.556955 5.033479
3  0.1332581  2.97334172 5.262912 8.236254
4  0.8064055  2.44649336 6.605434 9.051927
5 -1.5200194 -0.08463335 4.279925 4.195292

$dog
            a        b        c         d
1  0.73350259 1.133349 5.223977  6.357326
2  0.40957903 1.431519 5.332924  6.764442
3 -0.92179742 3.778318 6.924693 10.703011
4  1.92143657 1.055262 4.318544  5.373806
5 -0.07847994 2.045227 6.311894  8.357121

Upvotes: 3

Remko Duursma
Remko Duursma

Reputation: 2821

lapply works fine here. Note that a return(x) is needed here, otherwise we would just return the new vector.

res <- lapply(ls.1, function(x){
 x$d <- x$b + x$c
 return(x)
})

Upvotes: 13

thelatemail
thelatemail

Reputation: 93813

lapply your function over each list item. The function should add the two columns and then return the changed object:

my.fun <- function(data,x,y,out) {data[out] <- data[[x]] + data[[y]]; data}
lapply(ls.1, my.fun, x="b", y="c", out="d")

#$cat
#           a         b        c        d
#1 -0.3762835 0.8775672 3.599283 4.476851
#2  1.4728001 2.2384841 3.093956 5.332440
#...
#
#$dog
#           a         b        c        d
#1 -0.7669203 2.8219880 4.284088 7.106076
#2  0.4413280 1.1619882 4.440192 5.602180
#...

Upvotes: 4

Related Questions