[R] how to find "first" or "last" record after sort in R

[Bert Gunter]

Many ways to do this, of course, but if I understand correctly ?rle
may be the simplest, because you already have the data sorted by ID.

The following little example should give you the idea. It gets the
index of the last row in each id,, which you can then use to assign
NA's or whatever:

> id <- c(1,2,2,2,3,4,5,5)
> last.index <- cumsum(rle(test)$lengths)
> last.index
[1] 1 4 5 6 8

Cheers,
Bert


Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )

On Thu, Sep 9, 2021 at 12:00 PM Kai Yang via R-help
<r-help using r-project.org> wrote:
>
> Hello List,
> Please look at the sample data frame below:
>
> ID         date1              date2             date3
> 1    2015-10-08    2015-12-17    2015-07-23
>
> 2    2016-01-16    NA                 2015-10-08
> 3    2016-08-01    NA                 2017-01-10
> 3    2017-01-10    NA                 2016-01-16
> 4    2016-01-19    2016-02-24   2016-08-01
> 5    2016-03-01    2016-03-10   2016-01-19
> This data frame was sorted by ID and date1. I need to set the column date3 as missing for the "last" record for each ID. In the sample data set, the ID 1, 2, 4 and 5 has one row only, so they can be consider as first and last records. the data3 can be set as missing. But the ID 3 has 2 rows. Since I sorted the data by ID and date1, the ID=3 and date1=2017-01-10 should be the last record only. I need to set date3=NA for this row only.
>
> the question is, how can I identify the "last" record and set it as NA in date3 column.
> Thank you,
> Kai
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.