[R] aggregate with non-scalar functions

[carslaw]

Hi R-users,

Since R.2.11 aggregate can now deal with non-scalar functions, which is very
useful to me.

However, I have a question about how best to process the output.

test <- data.frame(a = rep(c("g1", "g2"), each = 50), b = runif(100))

res <- aggregate(test$b, list(group = test$a), function(x) quantile(x, probs
= c(0.05, 0.95)))

> res
  group       x.5%      x.95%
1    g1 0.00899229 0.91327509
2    g2 0.01863110 0.86187829

Looks like exactly what I want, but:

dim(res)
[1] 2 2

> res[, 2]
             5%       95%
[1,] 0.00899229 0.91327509
[2,] 0.01863110 0.86187829

> sapply(res, class)
   group        x 
"factor" "matrix" 

I first thought the output contained 3 columns, but it does not (and I
*think*
the help says as much).

So if I want 3 columns and all outputs accessible, I do:

res <- data.frame(subset(res, select = -x), res$x)
> dim(res)
[1] 2 3

And that works fine.

Is this the best way of obtaining a data frame with the 3 columns?

I know in this example I could have used tapply, but the actual data
consists of several variables.

Thanks.

David
-- 
View this message in context: http://r.789695.n4.nabble.com/aggregate-with-non-scalar-functions-tp2313759p2313759.html
Sent from the R help mailing list archive at Nabble.com.