R multiple statistics for multiple columns with data.table [duplicate]

Question

I want the same results as in R summarizing multiple columns with data.table but for several summary functions.

Here is an example

data <- as.data.table(list(x1 = runif(200), x2 = 10*runif(200), group = factor(sample(letters[1:2]))))

res <- data[, rbindlist(lapply(.SD, function(x) {
              return(list(name = "varname", mean = mean(x), sd = sd(x)))
           }))
          , by = group, .SDcols = c("x1", "x2")
          ]

And get the following result:

   group    name      mean        sd
1:     b varname 0.5755798 0.2723767
2:     b varname 5.5108886 2.7649262
3:     a varname 0.4906111 0.3060961
4:     a varname 4.7780189 2.9740149

How can I get column names ('x1', 'x2') in second column? I guess that I need to substitute rbindlist to something else, but what? Is there any simple solution?

Answer 1

An alternative way would be to construct your own function so that you can avoid this rbindlist wrap (which I find is unnecessary) which gives you the freedom of constructing your function the way you want:

tmp <- function(x) { 
    mm <- colMeans(x)
    ss=sapply(x, sd)
    list(names=names(x), mean=mm, sd=ss)
}

data[, tmp(.SD), by=group]
   group names      mean        sd
1:     a    x1 0.4988514 0.2770122
2:     b    x1 0.5246786 0.3014248
3:     a    x2 4.8031253 2.7978401
4:     b    x2 4.9104108 2.9135656

Answer 2

You can iterate your lapply on names(.SD) instead of .SD. Something like this :

data <- as.data.table(list(x1 = runif(200), x2 = 10*runif(200), group = factor(sample(letters[1:2]))))
res <- data[, rbindlist(lapply(names(.SD), function(name) {
              return(list(name = name, mean = mean(.SD[[name]]), sd = sd(.SD[[name]])))
           }))
          , by = group, .SDcols = c("x1", "x2")]

Which gives :

   group name      mean        sd
1:     b   x1 0.5344272 0.2697610
2:     b   x2 4.7628178 2.8313825
3:     a   x1 0.5008916 0.2686017
4:     a   x2 4.6175027 2.8942875