Consider the following:
> df <- data.frame(x = letters[1:15], y = rep(1:3, 5), z = rep(1:5, 3))
> df
x y z
1 a 1 1
2 b 2 2
3 c 3 3
4 d 1 4
5 e 2 5
6 f 3 1
7 g 1 2
8 h 2 3
9 i 3 4
10 j 1 5
11 k 2 1
12 l 3 2
13 m 1 3
14 n 2 4
15 o 3 5
I have another data frame, df2, say
> df2 <- data.frame(y = c(2, 3), z = c(2, 5))
> df2
y z
1 2 2
2 3 5
I would like to filter out the rows of df with y and z values as in df2. That is, the output should be something like
x y z
1 a 1 1
3 c 3 3
4 d 1 4
5 e 2 5
6 f 3 1
7 g 1 2
8 h 2 3
9 i 3 4
10 j 1 5
11 k 2 1
12 l 3 2
13 m 1 3
14 n 2 4
The pairing (y, z) is important. I've tried doing something like
df[!((df$y %in% df2$y) & (df$z %in% df2$z),]
but here's the problem: if I did this, not only would the pairs (2, 2) and (3, 5) be filtered, but (3, 2) and (2, 5) as well, which I do not want to happen.
Obviously, I could concatenate the columns and filter based on that, but I'm wondering if there's a better way to deal with this problem.
