determining which id's have a time point that's earlier than another time point in R-CodePudding

I am trying to find a way to return subject_ids with time_2 that's earlier than time_1

subject_id	Time_1	Time_2
191-5467	18:00	20:00
191-6784	18:30	18:50
191-3457	19:00	21:45
191-0987	19:30	20:00
191-1245	19:45	19:00
191-2365	20:00	19:00

So in this case the code would return subject id's 191-1245 and 191-2365 because both their time_2 is earlier than time_1.

CodePudding user response：

A dplyr solution:

library(dplyr)
df %>% 
  filter(strptime(Time_2, '%H:%M') < strptime(Time_1, '%H:%M')) %>% 
  pull(subject_id)
[1] "191-1245" "191-2365"

CodePudding user response：

We can convert to datetime with POSIXct and do the subset

subset(df1, as.POSIXct(Time_2, format = "%H:%M") < 
    as.POSIXct(Time_1, format = "%H:%M") )$subject_id
[1] "191-1245" "191-2365"

Or using data.table

library(data.table)
setDT(df1)[, subject_id[as.ITime(Time_2) < as.ITime(Time_1)]]
[1] "191-1245" "191-2365"

data

df1 <- structure(list(subject_id = c("191-5467", "191-6784", "191-3457", 
"191-0987", "191-1245", "191-2365"), Time_1 = c("18:00", "18:30", 
"19:00", "19:30", "19:45", "20:00"), Time_2 = c("20:00", "18:50", 
"21:45", "20:00", "19:00", "19:00")), class = "data.frame", row.names = c(NA, 
-6L))

CodePudding user response：

Similar to the @Jilber Urbina answer, but using lubridate.

library(lubridate)
library(dplyr)

df1 |>
  # Transform Time_1 and Time_2 to hour minutes with hm
  mutate(across(starts_with("Time"), hm)) |>
  # Filter to condition of interest
  filter(Time_2 < Time_1) |>
  # Pull the column with the id values
  pull(subject_id)