如何计算一栏内的份额答案

【问题标题】：How to calculate share within one column如何计算一栏内的份额
【发布时间】：2017-07-13 15:35:46
【问题描述】：

我有一个结构如下的数据框：

df <- structure(list(name1 =  c("A","A","B","B","A","A","B","B"), 
                 name2     =  c("B","B","C","C","ALL","ALL","ALL","ALL"),
                 pair_id   =  c(1,1,2,2,3,3,4,4),
                 year      =  c(2010, 2011, 2010, 2011, 2010, 2011,2010, 2011),
                 var1      =  c(1.5,2,4,5,12,15,20,18)), 
             .Names        =  c("name1","name2","pair_id","year", "var1"), 
            row.names      =  c("1", "2", "3", "4", "5", "6", "7", "8"), class =("data.frame"))

我想计算每年 var1（分母为 name2=ALL）和 pair_id 的百分比份额。输出应如下所示：

df <- structure(list(name1   =  c("A","A","B","B","A","A","B","B"), 
                 name2       =  c("B","B","C","C","ALL","ALL","ALL","ALL"),
                 pair_id     =  c(1,1,2,2,3,3,4,4),
                 year        =  c(2010, 2011, 2010, 2011,2010,2011,2010,2011),
                 var1        =  c(1.5,2,4,5,12,15,18,20), 
                 var1_share  =  c(0.125,0.133333,0.2,0.2777,1,1,1,1)), 
            .Names           =  c("name1","name2","pair_id","year", "var1","var1_share"), 
            row.names        =  c("1", "2", "3", "4", "5", "6", "7", "8"), class =("data.frame"))

提前谢谢你！

【问题讨论】：

嗨，Korone，你说的份额是什么意思？你有什么努力做到这一点？看来你已经计算了一年，所以你只需要为每一年和pair_id做那个函数。 combn 可以选择为每个组合执行一个功能
嗨@Llopis。是的，确实是百分比。我已经编辑了帖子以使其更清晰。

标签： r dataframe dplyr

【解决方案1】：

dplyr 解决方案：

df %>%
  group_by(name1, year) %>%
  mutate(denom = var1[name2 == "ALL"]) %>%
  mutate(var1_share = var1/denom)
# # A tibble: 8 x 7
# # Groups:   name1, year [4]
#   name1 name2 pair_id  year  var1 denom var1_share
#   <chr> <chr>   <dbl> <dbl> <dbl> <dbl>      <dbl>
# 1     A     B       1  2010   1.5    12  0.1250000
# 2     A     B       1  2011   2.0    15  0.1333333
# 3     B     C       2  2010   4.0    20  0.2000000
# 4     B     C       2  2011   5.0    18  0.2777778
# 5     A   ALL       3  2010  12.0    12  1.0000000
# 6     A   ALL       3  2011  15.0    15  1.0000000
# 7     B   ALL       4  2010  20.0    20  1.0000000
# 8     B   ALL       4  2011  18.0    18  1.0000000

【讨论】：