Max date in R column with sapply

问题:

I am trying to use sapply to get the max date in a column but it is returning a number instead of a date. Any idea how to resolve this? I can't seem to figure out why this is occurring..

 mtcars$datecolm = '2015-03-03'
 mtcars$datecolm[1] = '2015-09-09'
  mtcars$datecolm = as.Date(mtcars$datecolm)

    sapply(mtcars, max)   # why is it returning a number instead of a date??
    max(mtcars$datecolm)   # works correctly

Please use sapply given the way I have set this up... I know this works with apply(mtcars,2,max).

回答1:

We need to use lapply instead of sapply

lapply(mtcars, max)

as sapply returns a vector because of the simplify=TRUE default argument and vector can hold only a single class. As there are numeric columns, the 'Date' column (which is stored as integer) gets coerced to integer value. However, we can still use sapply if we use simplify = FALSE to return a list.

sapply(mtcars, max, simplify = FALSE)

Regarding the use of apply

apply(mtcars,2,max)

I would not recommend using apply as this will convert to matrix and matrix can hold only a single class. Here, it will be all character output as the 'Date' class got converted to character.

---

Regarding the problem mentioned by OP in the comments, when we have output that belong to different class, we could either use list or data.frame (as data.frame is a list) as c i.e. concatenate returns a vector and as mentioned above, it can hold only a single class.

do.call(rbind, lapply(mtcars, function(x) 
          data.frame(Class=class(x), Max=max(x, na.rm=TRUE))))