Split dataframe by index in columns of another dataframe

Question

I have created an index in another dataframe called "index"

I1 I2 I3 ...
1  3  5
2  4  6 
3  5  7

I would like to subset another dataframe like so, and store each result as a separate dataframe

column_name_from_index <- main_df[index[,i], ]

So the end result is I get main_df indexed by columns of index, output being dataframes I1, I2, I3 and so on. I am getting stuck on splitting out my results (I am not grabbing column name at the moment, but I would like advice on how to do that).

s <- seq(1,30)
df <- main_df[FALSE, ] #creates an empty dataframe into which I can put all columns from main_df.

for(i in s){
df <- main_df[index[,i], ]
df[i, ]
}

Anyway, it is always good to give a [reproducible example](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example/5963610). In your case, including example data for `main_df` (or the output of `dput(main_df)`) and the desired output would greatly benefit your question. — Jaap, Oct 11 '17 at 15:28

score 0 · Accepted Answer · answered Oct 11 '17 at 15:29

This will be more wieldy if you use a list instead of separate data frames. As Jaap shows, lapply is easy for this, and we can name the result:

index = data.frame(I1 = 1:3, I2 = 3:5)
main_df = data.frame(x = 1:6, y = 2:7)

result = lapply(index, function(i) main_df[i, ])
names(result) = names(index)
result
# $I1
#   x y
# 1 1 2
# 2 2 3
# 3 3 4
# 
# $I2
#   x y
# 3 3 4
# 4 4 5
# 5 5 6

You can access the individual data frames by number or name, e.g., result[["I2"]] or result[[2]]. See How to make a list of data frames? for a long discussion about why this is better than individual data frames (but also tips for converting the list to individual data frames for the exceptionally stubborn or masochistic).

Split dataframe by index in columns of another dataframe

1 Answers1