Question

I have some data that looks like the following. It is grouped by variable “Year” and I want to extract the percentiles of each observation of Score, with respect to the Year it is from, preferably as a vector.

Year   Score
2001   89
2001   70
2001   72
2001   ...
..........
2004   87
2004   90

etc.

How can I do this? aggregate will not work, and I do not think apply will work either.

Answer 1

Following up on Vince’s solution, you can also do this with plyr or by:

ddply(df, .(years), function(x) transform(x, percentile=ecdf(x$scores)(x$scores)))