CODEWITHSUNDEEP
rloopsif-statementmacrossubstr
BPeif
BPeif

Reputation: 370

How can i create a variable with the name of the substr i am looking for?

I am looking for a sort a macro that i could do under R, to create a variable from another one. I want to know if in the ZZZ variable, the string starts with a certain 3 first characters (by using the substr function). And so i would name the created variable by the name of the string that i am looking for. Some example to be more explicit :

My data is like:

ZZZ
M41
M410
M42
F128

I would like to create a loop like this:

data$xxx = ifelse(substr(data$ZZZ,1,3)=="xxx",1 ,0)

And so the data after would look like :

ZZZ M41 F12
M41 1 0
M410 1 0
M42 0 0
F128 0 1

Because i must do it for more than a hundred times, so creating a loop would help to decrease the size of the program.

I don't know if we can be able to do this type of macros under R.

Thanks for reading, BP

Upvotes: 0

Views: 191

Answers (1)

G. Grothendieck
G. Grothendieck

Reputation: 269481

1) Assuming the inputs in the Note at the end sapply over the target values.

cbind(dat, +sapply(target, `==`, substr(dat$ZZZ, 1, 3)))

giving:

   ZZZ M41 F12
1  M41   1   0
2 M410   1   0
3  M42   0   0
4 F128   0   1

2) If every element of ZZZ is to be used in a target then use target <- unique(substr(dat$ZZZ, 1, 3)) instead of the value for target shown in the Note.

3) If every element of ZZZ is to be used in a target then another approach is:

res <- cbind(dat, model.matrix(~ substr(dat$ZZZ, 1, 3) + 0, dat))
names(res) <- sub(".*\\)", "", names(res))
res

giving:

   ZZZ F12 M41 M42
1  M41   0   1   0
2 M410   0   1   0
3  M42   0   0   1
4 F128   1   0   0

Note

target <- c("M41", "F12")
dat <- structure(list(ZZZ = c("M41", "M410", "M42", "F128")),
  class = "data.frame", row.names = c(NA, -4L))

Upvotes: 1

Related Questions