Create a new variable from existing variable in dataframe using condition

Question

I have a big data frame like the example below.

          df <- data.frame(IND= seq(1:20), S = LETTERS[1:20], FA=c(0,0,133,0,2,2,2,0,0,4,4,4,6,6,0,0,0,4,2,8),
 MO = c(77,0,77,1,3,1,1,1,0,3,1,5,5,3,0,0,100,3,5,5)
 )

I need create two new variables (SFA and SMO) when the FA and MO are equals IND . I need the below output

                   out<- df <- data.frame(IND= seq(1:20), 
                   S = LETTERS[1:20], 
                   FA=c(0,0,133,0,2,2,2,0,0,4,4,4,6,6,0,0,0,4,2,8), 
                   MO = c(77,0,77,1,3,1,1,1,0,3,1,5,5,3,0,0,100,3,5,5), 
                               SFA=c(0,0,133,0,"B","B","B",0,0,"D","D","D","F","F",0,0,0,"D","B","H"),
 SMO=c(77,0,77,"A","C","A","A","A",0,"C","A","E","E","C",0,0,100,"C","E","E"))

I tried match the variables and after merge, but did not work very well.

THanks

akuiper · Accepted Answer · 2016-09-08 16:08:01Z

1

To pick up corresponding values from S where FA(MO) == IND, you can use match function to find out the index and subset from S as S[match(FA, IND)](S[match(MO, IND)]) and then use coalesce function to fill NAs in the match result with values from original vectors:

library(dplyr)
df %>% mutate(SFA = coalesce(as.character(S[match(FA, IND)]), as.character(FA)), 
              SMO = coalesce(as.character(S[match(MO, IND)]), as.character(MO)))

#   IND S  FA  MO SFA SMO
#1    1 A   0  77   0  77
#2    2 B   0   0   0   0
#3    3 C 133  77 133  77
#4    4 D   0   1   0   A
#5    5 E   2   3   B   C
#6    6 F   2   1   B   A
#7    7 G   2   1   B   A
#8    8 H   0   1   0   A
# ...

answered Sep 8, 2016 at 16:08

akuiper

216k33 gold badges362 silver badges379 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Create a new variable from existing variable in dataframe using condition

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related