如何计算 Dplyr 中每个组的分享率？

Question

I have a data set that shows the number of products for each group and shop.

df <- tribble(
  ~shop_id, ~group, ~group_2, ~products,  
    '1',      'A',     'Z',      10,                
    '2',      'B',     'Y',      20, 
    '3',      'A',     'X',      30, 
    '4',      'B',     'X',      40, 
    '5',      'A',     'R',      10
)

I now want to see the share of products for each shop id and group. But I want to exclude the group 2 column in the data. For instance, there are 50 products in group A, so the share for shop 1 should be 0.2. Here is the desired output:

df <- tribble(
  ~shop_id, ~group, ~products,  ~share_products, 
  '1',      'A',       10,            0.2,    
  '2',      'B',       20,            0.33,
  '3',      'A',       30,            0.6,
  '4',      'B',       40,            0.66,
  '5',      'A',       10,            0.2
)

How can I do this?

Answer 1

After grouping, divide by the sum of 'products'

library(dplyr)
df1 <- df %>% 
 group_by(group) %>% 
 mutate(share_products = products/sum(products)) %>% 
 ungroup

-output

df1
# A tibble: 5 × 4
  shop_id group products share_products
  <chr>   <chr>    <dbl>          <dbl>
1 1       A           10          0.2  
2 2       B           20          0.333
3 3       A           30          0.6  
4 4       B           40          0.667
5 5       A           10          0.2  

---

If there are several 'products' column, loop across those columns to create the corresponding 'share' columns

df1 <- df %>%
      group_by(group) %>%
      mutate(across(contains('products'),  ~.x/sum(.x),
       .names = "share_{.col}")) %>%
      ungroup

Answer 2

We could use prop.table after grouping

df %>%
  select(-group_2) %>% 
  group_by(group) %>%
  mutate(prop = prop.table(products))

  shop_id group products  prop
  <chr>   <chr>    <dbl> <dbl>
1 1       A           10 0.2  
2 2       B           20 0.333
3 3       A           30 0.6  
4 4       B           40 0.667
5 5       A           10 0.2