CODEWITHSUNDEEP
rlist
user1987607
user1987607

Reputation: 2157

R - list of dataframes - calculate and add new rows

I have the following list of dataframes

d1 <- data.frame(var1 = 10, var2 = 20, var3 = 30)
d2 <- data.frame(var4 = 40, var5 = 50, var6 = 60)
my.list <- list(d1, d2)

> my.list
[[1]]
    var1 var2 var3
  1   10   20   30

[[2]]
    var4 var5 var6
  1   40   50   60

Now I want to add new rows to each of these dataframes that contain the following content:

values <- c(0.75, 0.5, 0.25)

d1$var1 * values[1] #new second row d1
d1$var1 * values[2] #new third row d1
d1$var1 * values[3] #new fourth row d1

This needs to be done for all $var variables in each dataframe and all dataframes in the list.

the new d1 would look like this:

    var1 var2 var3
  1   10   20   30
  2  7.5   15 22.5
  3    5   10   15
  4  2.5    5  7.5

How could I achieve this?

Upvotes: 3

Views: 599

Answers (3)

Cole
Cole

Reputation: 11255

Here's an interesting use of rapply(). 

rapply(my.list, function(x) c(x,x*values), how = 'list')

[[1]]
[[1]]$var1
[1] 10.0  7.5  5.0  2.5

[[1]]$var2
[1] 20 15 10  5

[[1]]$var3
[1] 30.0 22.5 15.0  7.5


[[2]]
[[2]]$var4
[1] 40 30 20 10

[[2]]$var5
[1] 50.0 37.5 25.0 12.5

[[2]]$var6
[1] 60 45 30 15

You can get it back to your intended output by wrapping it in lapply(..., data.frame)

lapply(rapply(my.list, function(x) c(x,x*values), how = 'list' ), data.frame)

[[1]]
  var1 var2 var3
1 10.0   20 30.0
2  7.5   15 22.5
3  5.0   10 15.0
4  2.5    5  7.5

[[2]]
  var4 var5 var6
1   40 50.0   60
2   30 37.5   45
3   20 25.0   30
4   10 12.5   15

The rapply() itself is faster but it loses all advantages after the result is coerced back to a data.frame.

Upvotes: 0

NelsonGon
NelsonGon

Reputation: 13319

Almost similar to @akrun 's, we can do(might be more computationally expensive and also has the disadvantage of forming a matrix) :

lapply(my.list,sapply,function(x) append(x,
                                do.call(`*`,list(x,values))))
[[1]]
     var1 var2 var3
[1,] 10.0   20 30.0
[2,]  7.5   15 22.5
[3,]  5.0   10 15.0
[4,]  2.5    5  7.5

[[2]]
     var4 var5 var6
[1,]   40 50.0   60
[2,]   30 37.5   45
[3,]   20 25.0   30
[4,]   10 12.5   15

Upvotes: 2

akrun
akrun

Reputation: 887128

An option would be to loop over the list and multiply each column with the 'values' vector and rbind the original data row with the new row

lapply(my.list, function(x) rbind(x, sapply(x, `*`, values)))
#[[1]]
#  var1 var2 var3
#1 10.0   20 30.0
#2  7.5   15 22.5
#3  5.0   10 15.0
#4  2.5    5  7.5

#[[2]]
#  var4 var5 var6
#1   40 50.0   60
#2   30 37.5   45
#3   20 25.0   30
#4   10 12.5   15

Or an option with tidyverse using add_row and map

library(tidyverse)
map(my.list, ~  .x %>%
                add_row(!!! map(., `*`, values)))
#[[1]]
#  var1 var2 var3
#1 10.0   20 30.0
#2  7.5   15 22.5
#3  5.0   10 15.0
#4  2.5    5  7.5

#[[2]]
#  var4 var5 var6
#1   40 50.0   60
#2   30 37.5   45
#3   20 25.0   30
#4   10 12.5   15

Upvotes: 5

Related Questions