features for missing data

[njtierney]

features for missing data

[njtierney]

Perhaps something like this?

library(naniar)
library(brolgar)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

wages %>%
  group_by(id) %>%
  miss_case_summary() %>%
  filter(n_miss > 0) %>%
  arrange(-n_miss)
#> # A tibble: 799 x 4
#>       id  case n_miss pct_miss
#>    <int> <int>  <int>    <dbl>
#>  1  2766     6      2    14.3 
#>  2  2786     3      2    14.3 
#>  3  3340     3      2    14.3 
#>  4  4068     7      2    14.3 
#>  5  6439     8      2    14.3 
#>  6  6762     9      2    14.3 
#>  7  9093     2      2    14.3 
#>  8  9436     2      2    14.3 
#>  9 10330     6      2    14.3 
#> 10    53     5      1     7.14
#> # … with 789 more rows

wages %>%
  group_by(id) %>%
  summarise_at(.vars = vars(lnw),
               .funs = list(n_miss = function(x) sum(is.na(x)))) %>%
  arrange(-n_miss)
#> # A tibble: 888 x 2
#>       id n_miss
#>    <int>  <int>
#>  1    31      0
#>  2    36      0
#>  3    53      0
#>  4   122      0
#>  5   134      0
#>  6   145      0
#>  7   155      0
#>  8   173      0
#>  9   206      0
#> 10   207      0
#> # … with 878 more rows

^{Created on 2019-04-09 by the reprex package (v0.2.1)}

[njtierney]

• Aim to move any new functions from this into naniar.
• Aim to demonstrate missingness functions in a workflow