Question

Possible Duplicate:
faster way to create variable that aggregates a column by id

So the thing is that i have following data loaded from a CSV file:

id      value2  value3
1.000   0.010   14        
1.000   0.019   15        
0.995   0.024   13        
0.995   0.031   20        
0.990   0.012   13        
.....

I want to calculate the mean/median etc. value of value2 and value3 in groups of id. Afterwards the plan was to be able to sort the result by either value2 or value3.

Is there a way to do such task?

Thanks in advance.

Answer 1

library(plyr)
result <- ddply(df, .(id), function(x) { 
    data.frame(mv2 = mean(x$value2), mv3 = mean(x$value3))
    })

# order by mean value2
arrange(result, mv2)
# and for value 3
arrange(result, mv3)