[R] Need help with merge
analyst41 at hotmail.com
analyst41 at hotmail.com
Thu Feb 10 03:16:30 CET 2011
Have
> actualsdf
ID Name datadate val
1 23 Acme Corp 1 23
2 23 Acme Corp 2 43
3 23 Acme Corp 3 54
4 23 Acme Corp 4 65
5 23 Acme Corp 5 23
6 23 Acme Corp 6 43
7 23 Acme Corp 7 NA
8 23 Acme Corp 8 43
9 23 Acme Corp 9 54
10 23 Acme Corp 10 32
> fcstdf
fcstrundate fcstdate fcst ID Name
1 5 6 22 23 Acme Corp
2 6 7 43 23 Acme Corp
3 7 8 54 23 Acme Corp
4 8 9 23 23 Acme Corp
5 9 10 NA 23 Acme Corp
6 10 11 13 23 Acme Corp
> mergeddf =merge(fcstdf,actualsdf,by.x = "fcstdate",by.y = "datadate",all =TRUE)
> mergeddf
fcstdate fcstrundate fcst ID.x Name.x ID.y Name.y val
1 1 NA NA NA <NA> 23 Acme Corp 23
2 2 NA NA NA <NA> 23 Acme Corp 43
3 3 NA NA NA <NA> 23 Acme Corp 54
4 4 NA NA NA <NA> 23 Acme Corp 65
5 5 NA NA NA <NA> 23 Acme Corp 23
6 6 5 22 23 Acme Corp 23 Acme Corp 43
7 7 6 43 23 Acme Corp 23 Acme Corp NA
8 8 7 54 23 Acme Corp 23 Acme Corp 43
9 9 8 23 23 Acme Corp 23 Acme Corp 54
10 10 9 NA 23 Acme Corp 23 Acme Corp 32
11 11 10 13 23 Acme Corp NA <NA> NA
I would like mergeddf to look like
> cleanmergeddf
fcstdate fcstrundate fcst val ID Name
1 1 NA NA 23 23 Acme Corp
2 2 NA NA 43 23 Acme Corp
3 3 NA NA 54 23 Acme Corp
4 4 NA NA 65 23 Acme Corp
5 5 NA NA 23 23 Acme Corp
6 6 5 22 43 23 Acme Corp
7 7 6 43 NA 23 Acme Corp
8 8 7 54 43 23 Acme Corp
9 9 8 23 54 23 Acme Corp
10 10 9 NA 32 23 Acme Corp
11 11 10 13 NA 23 Acme Corp
I can think of an awkward way - but is there a direct merged command
that would produce the final output?
Thanks.
More information about the R-help
mailing list