How to use dplyr as alternative to aggregate

Question

I have a dataframe times that looks like this:

user     time
A        7/7/2010
B        7/12/2010
C        7/12/2010
A        7/12/2010 
C        7/         


        

Answer 1

Based on the dplyr solution by eipi10 and the suggestion of nrussell, I've written the following solution using data.table.

First you need to format the variable times:

times$time = as.Date(times$time, "%m/%d/%Y")

Then you'll need to convert times to a data.table using:

library(data.table)
times <- as.data.table(times)

Overwriting times was useful for my purposes but you may want to instantiate a new variable. After formatting your dataframe as a data.table just do:

new.times <- times[, 
                    .(first = min(time),
                      last = max(time),
                      n = .N,
                      meandiff = mean(diff(time)),
                      mindiff = min(diff(time)),
                      numdiffuniq = length(unique(diff(time))),
                      by='user')]

Running on a linux virtual machine with 128G RAM and using a sample of 1000 entires, the elapsed runtime was 0.43s.

See this tutorial for more on data.table.