Subsetting a vector using another boolean vector in R

Question

Using the following two R vectors, I want to extract a subset of valMe using the boolean values in boolMe. In addition, I would like to have two possible outputs, one where the FALSE values in boolMe are ommited from valMe, and one where the FALSE values are replaced by NA. Further illustration of what I want to do in code:

Input

boolMe<-c(FALSE, TRUE, TRUE, TRUE, FALSE, TRUE)
valMe<-1:6

Intended output

NA 2 3 4 NA 6

or

2 3 4 6

In python, I would do it something like this: `>>> a=[False, True, False, False, False, False] >>> b=[10, 11, 12, 13, 14, 15] >>> [bb for (aa,bb) in zip(a,b) if aa] [11]` — hhh, Dec 26 '11 at 04:49
In `python3`, `import itertools; list(itertools.compress(b ,a))` — kev, Dec 26 '11 at 05:11

score 14 · Accepted Answer · answered Dec 26 '11 at 05:01

14

You can directly index valMe by using the [ operator:

> valMe[boolMe]
[1] 2 3 4 6

See section 2.7 of the intro manual for more detail.

answered Dec 26 '11 at 05:01

Chase

67,710
18
144
161

How would you do it if you have a matrix instead and you want to keep all columns? – confused Jan 08 '20 at 20:13

Ari B. Friedman · Answer 2 · 2011-12-26T17:56:39.380

8

Similarly, if you want the NAs:

> valMe[!boolMe] <- NA
> valMe
[1] NA  2  3  4 NA  6

The ! negates the logical boolean so you select the values you want missing. Then, in a touch of R awesomeness, you assign NA to the selected values.

edited Dec 26 '11 at 17:56

answered Dec 26 '11 at 14:31

Ari B. Friedman

71,271
35
175
235

Subsetting a vector using another boolean vector in R

2 Answers2