如何使用mutate_at获取百分位数？

Question

Consider this simple example

> tibble(var1 = c(1,2,3,4,5),
+        boo1 = c(1,2,3,4,5))
# A tibble: 5 x 2
   var1  boo1
  <dbl> <dbl>
1     1     1
2     2     2
3     3     3
4     4     4
5     5     5

Here I want to get the current value expressed as a percentage using ecdf . This is expected to work but it does not.

> tibble(var1 = c(1,2,3,4,5),
+        boo1 = c(1,2,3,4,5)) %>% 
+   mutate_at(vars(contains('boo')), 
+             .funs = funs(ecdf(.)(.)))
Error in mutate_impl(.data, dots) : 
  Evaluation error: 'x' and 'y' lengths differ.

Instead, this works

> tibble(var1 = c(1,2,3,4,5),
+        boo1 = c(1,2,3,4,5)) %>% mutate(percentile = ecdf(boo1)(boo1))
# A tibble: 5 x 3
   var1  boo1 percentile
  <dbl> <dbl>      <dbl>
1     1     1        0.2
2     2     2        0.4
3     3     3        0.6
4     4     4        0.8
5     5     5        1  

What is the issue here? Thanks!

Answer 1

Two alternatives would be

ecdfEval <- function(x) ecdf(x)(x)
tbl %>% mutate_at(vars(contains('boo')), ecdfEval)
# A tibble: 5 x 2
#    var1  boo1
#   <dbl> <dbl>
# 1     1   0.2
# 2     2   0.4
# 3     3   0.6
# 4     4   0.8
# 5     5   1

and

tbl %>% mutate_at(vars(contains('boo')), funs(do.call(ecdf(.), list(.))))
# A tibble: 5 x 2
#    var1  boo1
#   <dbl> <dbl>
# 1     1   0.2
# 2     2   0.4
# 3     3   0.6
# 4     4   0.8
# 5     5   1

It is indeed strange that your approach didn't work, seems like using . in nested functions is a problem.

---

Edit (better option): It seems like it may be more about funs than mutate_at since, as @Nate noticed, we can indeed use simply

tbl %>% mutate_at(vars(contains('boo')), .funs = ~ecdf(.)(.))

(See @Nate's comment below for more details.)