Subsetting a data frame and replacing a column based on condition - r

Question

I am working on a data frame with three columns labelled as id, time1 and time2. A sample is:
df <-
structure(
list(
id = c(1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L),
time1 = c(12L, 5L, 3L, 5L, 6L, 30L, 3L, 30L, 7L, 2L, 17L, 5L, 8L, 3L, 22L, 5L, 15L, 4L, 7L, 23L),
time2=c(23L,23L,23L,23L,23L,22L,22L,22L,22L,22L,25L,25L,25L,25L,25L,24L,24L,24L,24L,24L)
),
.Names = c("id", "time1","time2"),
class = "data.frame",
row.names = c(NA,-20L)
)
I am using R and I am trying to subset this data and replace column time2 with a new column based on the following criteria:
Sum the values of time1 for each id until it is greater than or equal to the corresponding value of time2 for that id.
Replace the cells in time1 where the summations terminate with the corresponding time2 value for each id.
Column time2 is to be replaced with a new column labelled as status which consists of 0's and 1's. That is, status takes on 1 for the non-replaced values of time1 and 0 for all the replaced values of time1.
In summary, I am expecting to see something like this:
df <-
structure(
list(
id = c(1L, 1L, 1L, 1L, 2L, 3L, 3L, 3L, 4L, 4L, 4L),
time1 = c(12L, 5L, 3L, 23, 22L, 17L, 5L, 25L, 5L, 15L, 24L),
status=c(1L,1L,1L,0L,0L,1L,1L,0L,1L,1L,0L)
),
.Names = c("id", "time1","status"),
class = "data.frame",
row.names = c(NA,-11L)
)
I greatly appreciate any help on this.

Answer 1

We can do the following:
library(tidyverse);
df %>%
group_by(id) %>%
mutate(
status = as.numeric(cumsum(time1) < time2),
time1 = ifelse(status == 1, time1, time2)) %>%
group_by(id, status) %>%
mutate(n = 1:n()) %>%
ungroup() %>%
filter(status == 1 | (status == 0 & n == 1)) %>%
select(-n, -time2)
## A tibble: 11 x 3
# id time1 status
# <int> <int> <dbl>
# 1 1 12 1.
# 2 1 5 1.
# 3 1 3 1.
# 4 1 23 0.
# 5 2 22 0.
# 6 3 17 1.
# 7 3 5 1.
# 8 3 25 0.
# 9 4 5 1.
#10 4 15 1.
#11 4 24 0.
Explanation: We group rows by id, then calculate the cumulative sum of time1 entries, and flag those rows where cumsum(time1) < time2 with 1, else with 0; we replace time1 entries with time2 entries if status == 1. Lastly we need to remove excess status = 0 rows; to do so, we regroup by id and status, number rows consecutively, and keep only one row for status = 0 per id.