select values on multiple data frames by multiple groups

Question

I am trying to run a function that takes value1, value2, value3 of the same time and name from multiple data frames. output <- function(value1, value2, value3) For example:

DF1:
time       name      value1
Jan 1990      1      4
Feb 1990      1      2
Mar 1990      1      3
Jan 1990      2      2
Feb 1991      2      1
DF2:
time       name      value2
Jan 1990      1      4
Feb 1990      1      2
Mar 1990      1      3
Jan 1990      2      2
Feb 1991      2      1
DF3:
time       name      value3
Jan 1990      1      4
Feb 1990      1      2
Mar 1990      1      3
Jan 1990      2      2
Feb 1991      2      1

Should I use join and then apply on each row?

output
time       name    output 
Jan 1990      1      4
Feb 1990      1      2
Mar 1990      1      3
Jan 1990      2      2
Feb 1991      2      1

Vivek Kalyanarangan · Accepted Answer · 2019-01-17T06:30:28.467

0

Use -

Option 1: dplyr library(dplyr)

output <- inner_join(inner_join(df1, df2, by=c("time", "name")), 
                 df3, by=c("time", "name"))

Option 2: plyr

library(plyr)
join_all(list(df1,df2,df3), by=c('time','name'), type='inner')

Option 3: Base R merge+reduce

Use merge + reduce - credits to @akrun

Reduce(function(...) merge(..., by=c("time","name"), all.x=TRUE), list(df1,df2,df3))

Output

      time name value1
1 Jan 1990    1      4
2 Feb 1990    1      2
3 Mar 1990    1      3
4 Jan 1990    2      2
5 Feb 1991    2      1

You can then rename the columns as per your necessity.

edited Jan 17 '19 at 06:30

answered Jan 17 '19 at 05:45

Vivek Kalyanarangan

8,951
1
23
42

Thank you! But what if I have let's say 10 DFs. Is there a way of using `Reduce`? Also, I need all of value 1, value 2 and value 3. So I just rewrite `by = c("time", "name")`. Correct? – YellowRiver Jan 17 '19 at 06:15
@YellowRiver added more options pls check – Vivek Kalyanarangan Jan 17 '19 at 06:30

select values on multiple data frames by multiple groups

1 Answers1