How does sapply work in the example below?

Question

sapply(list(runif (10), runif (10)), 
       function(x) c(min = min(x), mean = mean(x), max = max(x)))

Why is the above code giving a matrix of 3 rows and 2 columns and not 2 rows and 3 columns ?

Answer 1

Edit (original answer with explanation below) :

We can use ldply from plyr as it directly gives us a data.frame object.

plyr::ldply(list(runif(10),runif(10)),
            function(x) c(Min=min(x),Mean=mean(x),Max=max(x)))


  #         Min      Mean       Max #No seed
   # 1 0.03964594 0.3960042 0.7345305
   # 2 0.05233872 0.3811633 0.9713190

Original:

It makes a list with two components:

list(runif(10),runif(10))

From each list, we find the minimum,maximum and mean hence the output. To understand why, try running this alone:

sapply(list(runif (10)), function(x) c(min = min(x), mean = mean(x), max = max(x)))

If you change the simplify argument, you can get the output you wish(I assume) to have:

sapply(list(runif (10)), function(x) c(min = min(x), mean = mean(x), max = max(x)),
       simplify = F)

 #[[1]]
 #         min      mean       max #No seed
 #  0.2482040   0.4845170 0.8294896 

NOTE:: You need to set.seed for reproducibility.