Question

I’ve got a data frame like this:

id date               amt
1  2012-05-03 10:33   32
2  2012-06-01 12:49   242
2  2012-06-05 00:09   43
3  2012-06-03 05:19   323
3  2012-06-08 08:45   12
4  2012-06-09 12:38   32
5  2012-06-09 10:31   53

Now I want to remove the duplicate id‘s so that the one with the earliest date is selected. The number of duplicate entries varies. I care only about the first occurrence of each particular id and the corresponding amt, all other entries should be removed.

I understand how I can do this with a loop but I feel there can be a short and elegant solution in R.

Answer 1

Try something like newdata <- data[!duplicated(data$id), ].

EDIT: As @Aaron and others have noted below, this assumes your data is sorted:

testdata <- data[with(data, order(id, date)), ]

newdata <- testdata[!duplicated(testdata$id), ]