We can convert the ‘DATE’ to Date
class, then do a group by ‘FID’ and slice
the row with the max
value in ‘DATE’
library(dplyr)
library(lubridate)
withrepeats %>%
mutate(DATE = mdy(DATE)) %>%
arrange(FID, desc(CID)) %>%
group_by(FID) %>%
mutate(mxDate = if(any(CT == 1)) DATE[which.max(DATE)] else
DATE[which.max(DATE)]) %>%
filter(DATE == mxDate & !duplicated(DATE)) %>%
ungroup %>%
slice(c(2, 4, 5, 1, 3)) %>%
select(-mxDate)
-output
# A tibble: 5 x 6
# FID ID CID CT DROPS DATE
# <int> <chr> <int> <int> <int> <date>
#1 123 CV 3 1 1 2020-11-03
#2 456 LO 2 1 1 2020-10-05
#3 678 IP 1 1 2 2020-11-02
#4 111 AK 1 2 2 2020-11-02
#5 222 PL 4 2 2 2020-11-01
data
withrepeats <- structure(list(FID = c(123L, 123L, 123L, 456L, 456L, 678L, 678L,
678L, 111L, 111L, 222L, 222L), ID = c("CV", "CV", "CV", "LO",
"LO", "IP", "IP", "IP", "AK", "AK", "PL", "PL"), CID = c(1L,
2L, 3L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 4L, 3L), CT = c(2L, 2L, 1L,
1L, 1L, 2L, 1L, 2L, 2L, 2L, 2L, 2L), DROPS = c(3L, 2L, 1L, 2L,
1L, 3L, 2L, 1L, 2L, 1L, 2L, 2L), DATE = c("11-3-2020", "11-3-2020",
"11-3-2020", "10-4-2020", "10-5-2020", "11-1-2020", "11-2-2020",
"10-29-2020", "11-2-2020", "11-1-2020", "11-1-2020", "11-1-2020"
)), class = "data.frame", row.names = c(NA, -12L))
CLICK HERE to find out more related problems solutions.