R 中的元素子集

Question

I am trying to print specific observations from a data frame. Consider this simple example:

df <- data.frame(ID = c(1,1,1,2,2,2,3,3),
                 Week = c(1,2,2,1,1,2,1,1),
                 Y = c(4,2,6,7,5,3,1,9))

I would like to (only) print the rows where (ID = 1 & Week = 2), (ID = 2 & Week = 1) as well as (ID = 3 & Week = 1), giving this output:

rbind(df[(df$ID == 1) & (df$Week == 2),],
      df[(df$ID == 2) & (df$Week == 1),],
      df[(df$ID == 3) & (df$Week == 1),])

The values to be used for indexing are stored in a vector for each variable:

IDidx <- c(1,2,3)
Weekidx <- c(2,1,1)

Is there any solution that takes these vectors and indexes element-wise from them as I have done it "manually" using rbind()?

Thanks for your help!

Answer 1

We can create a data frame based on IDidx and Weekidx , and then use the semi_join from the dplyr package.

inx <- data.frame(ID = IDidx, Week = Weekidx)

library(dplyr)

df %>% semi_join(inx, by = c("ID", "Week"))
#   ID Week Y
# 1  1    2 2
# 2  1    2 6
# 3  2    1 7
# 4  2    1 5
# 5  3    1 1
# 6  3    1 9