Is there R function for removing specific column condition

Question

Hello all my df looks like

PID V1
123 1
123 2
123 3
111 1
111 2
111 1
122 3
122 1
122 1
333 1
333 4
333 2

I want to delete rows contains 1 and 2 event alone for the PID

and expected output

PID V1
123 1
123 2
123 3
122 3
122 1
122 1
333 1
333 4
333 2

Answer 1

You can do this in base R :

subset(df, !ave(V1 %in% 1:2, PID, FUN = all))

#   PID V1
#1  123  1
#2  123  2
#3  123  3
#7  122  3
#8  122  1
#9  122  1
#10 333  1
#11 333  4
#12 333  2

dplyr

library(dplyr)
df %>% group_by(PID) %>% filter(!all(V1 %in% 1:2))

or data.table :

library(data.table)
setDT(df)[, .SD[!all(V1 %in% 1:2)], PID]

The logic of all of them is the same. Remove groups ( PID ) who have only 1 and 2 in V1 column.

data

df <- structure(list(PID = c(123L, 123L, 123L, 111L, 111L, 111L, 122L, 
122L, 122L, 333L, 333L, 333L), V1 = c(1L, 2L, 3L, 1L, 2L, 1L, 
3L, 1L, 1L, 1L, 4L, 2L)), class = "data.frame", row.names = c(NA, -12L))