基于级别的子集数据框

Question

I have the following dataframe in R. I want to subset it based on three criteria, for each unique value of x within each level of id

1. If there is only 1 value of x, keep that row
2. If x has the same value of z, with two different values of y, keep the row where y does not = 1.3
3. If x has three values of z, keep the two rows where y does not = 1.3

 id x y z a 1 0.2 100 a 2 1 200 a 2 1.3 200 b 1 0.5 400 b 1 1 500 b 1 1.3 600

the solution would look like this:

 id x y z a 1 0.2 100 a 2 1 200 b 1 0.5 400 b 1 1 500

Any help would be appreciated

Answer 1

We can group by 'id', 'x' and filter based on the conditions

library(dplyr)
df1 %>% 
   group_by(id, x) %>% 
   filter(n() == 1|(n() > 1 & y != 1.3))

data

df1 <- structure(list(id = c("a", "a", "a", "b", "b", "b"), x = c(1L, 
2L, 2L, 1L, 1L, 1L), y = c(0.2, 1, 1.3, 0.5, 1, 1.3), z = c(100L, 
200L, 200L, 400L, 500L, 600L)), class = "data.frame", row.names = c(NA, 
-6L))