How to match two columns of different data frame using R? Matched values should align at same position and missing values will be NAs.
Inputs:
df1$A: w x y
df2$B: x y z
Expected output data frame:
w x y NA
NA x y z
Asked
Active
Viewed 93 times
0
Gavin Simpson
- 170,508
- 25
- 396
- 453
feng63600
- 183
- 1
- 2
- 13
-
We can use `match` – akrun May 19 '16 at 06:04
-
`merge(df1, df2, by.x = 'A', by.y = 'B', all = TRUE)` ? – Jaap May 19 '16 at 06:47
-
Please read the info on [ask] and how to give a [reproducible example](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) – Jaap May 19 '16 at 07:14
2 Answers
1
There should be a better way to represent this but right now I can figure this out. As mentioned in comments by @akrun, you can use match
unique(rbind(cbind(as.character(df1$A), as.character(df2$B[match(df1$A, df2$B)])),
cbind(as.character(df2$B), as.character(df1$A[match(df2$B, df1$A)]))))
# [,1] [,2]
#[1,] "w" NA
#[2,] "x" "x"
#[3,] "y" "y"
#[4,] "z" NA
Ronak Shah
- 377,200
- 20
- 156
- 213
0
Well this works fine...
df1$B <- ifelse(df1$A %in% df2$B, as.character(df1$A), NA)
df3 <- merge(df1,df2,by='B',all.x = T,all.y = T)
Gaurav
- 1,597
- 2
- 14
- 31
-
I think the ifelse is not needed, `merge(df1, df2, by.x = 'A', by.y = 'B', all = TRUE)` will do just fine as far i can judge atm. – Jaap May 19 '16 at 06:49
-
@ProcrastinatusMaximus Well this just gives me (A: w x y z) as a single column dataframe output... Are you sure it works? – Gaurav May 19 '16 at 06:55
-