R 對列和行求和以進行配對分析

Question

一個有一個電子表格，用戶在其中輸入前后列中的年數。 一個例子如下所示。

有人想將 Yr_1、Yr_2、... Yr_n 行中提供的相應年份的數量相加，這些行對應於每個站點之前和之后的年份。

df = data.frame(
  SITE = c(1,2,3),
  Before=c(2,2,1),
  After=c(2,5,6),
  Yr_1=c(4,2,1),
  Yr_2=c(4,0,0),
  Yr_3=c(1,0,2),
  Yr_4=c(5,2,1),
  Yr_5=c(2,2,0),
  Yr_6=c(1,0,1),
  Yr_7=c(1,2,1))

df %>% mutate(
  Total_Counts_Before=

ifelse(
  Before == 1, rowSums(across(Yr_1)),
  ifelse(
    Before == 2, rowSums(across(Yr_1:Yr_2)),
    ifelse(
      Before == 3, rowSums(across(Yr_1:Yr_3)),
      ifelse(
        Before == 4, rowSums(across(Yr_1:Yr_4)),
        ifelse(
          Before == 5, rowSums(across(Yr_1:Yr_5)),
          ifelse(
            Before == 6, rowSums(across(Yr_1:Yr_6)),
            ifelse(
              Before == 7, rowSums(across(Yr_1:Yr_7)),"")))))))


)

但是，可能有多個組合（即 Before_Yr_n x After_Yr_n），因此需要在用戶表中為 After Year Counts 創建一組單獨的 7 列。

df = data.frame(
  SITE = c(1,2,3),
  Before=c(2,2,1),
  After=c(2,5,6),
  Before_Yr_1=c(4,2,1),
  Before_Yr_2=c(4,0,0),
  Before_Yr_3=c(0,0,0),
  Before_Yr_4=c(0,0,0),
  Before_Yr_5=c(0,0,0),
  Before_Yr_6=c(0,0,0),
  Before_Yr_7=c(0,0,0),
  
  After_Yr_1=c(1,0,0),
  After_Yr_2=c(5,2,2),
  After_Yr_3=c(2,2,1),
  After_Yr_4=c(1,0,0),
  After_Yr_5=c(1,2,1),
  After_Yr_6=c(0,0,1),
  After_Yr_7=c(0,0,0)
  
  )

df = df %>% mutate(
  Total_Counts_Before=

ifelse(
  Before == 1, rowSums(across(Before_Yr_1)),
  ifelse(
    Before == 2, rowSums(across(Before_Yr_1:Before_Yr_2)),
    ifelse(
      Before == 3, rowSums(across(Before_Yr_1:Before_Yr_3)),
      ifelse(
        Before == 4, rowSums(across(Before_Yr_1:Before_Yr_4)),
        ifelse(
          Before == 5, rowSums(across(Before_Yr_1:Before_Yr_5)),
          ifelse(
            Before == 6, rowSums(across(Before_Yr_1:Before_Yr_6)),
            ifelse(
              Before == 7, rowSums(across(Before_Yr_1:Before_Yr_7)),""))))))),
  
  Total_Counts_After=
ifelse(
  After == 1, rowSums(across(After_Yr_1)),
  ifelse(
    After == 2, rowSums(across(After_Yr_1:After_Yr_2)),
    ifelse(
      After == 3, rowSums(across(After_Yr_1:After_Yr_3)),
      ifelse(
        After == 4, rowSums(across(After_Yr_1:After_Yr_4)),
        ifelse(
          After == 5, rowSums(across(After_Yr_1:After_Yr_5)),
          ifelse(
            After == 6, rowSums(across(After_Yr_1:After_Yr_6)),
            ifelse(
              After == 7, rowSums(across(After_Yr_1:After_Yr_7)),"")))))))

)

有沒有一種方法可以避免將用戶輸入數據分別輸入 14 列或 3x14 矩陣？

所需的 output 可能看起來類似於，但必須包含 Total Before 和 After Count 列，如下所示：

df %>% kable()

地點	前	后	Before_Yr_1	Before_Yr_2	Before_Yr_3	Before_Yr_4	Before_Yr_5	Before_Yr_6	Before_Yr_7	After_Yr_1	After_Yr_2	After_Yr_3	After_Yr_4	After_Yr_5	After_Yr_6	After_Yr_7	Total_Counts_Before	Total_Counts_After
1	2	2	4	4	0	0	0	0	0	1	5	2	1	1	0	0	8	6
2	2	5	2	0	0	0	0	0	0	0	2	2	0	2	0	0	2	6
3	1	6	1	0	0	0	0	0	0	0	2	1	0	1	1	0	1	5

Answer 1

我建議您將數據更改為長格式，然后過濾您感興趣的結果，我不知道您的之后或之前的概念是否包括當前年份但您可以輕松更改它

library(tidyverse)

df = data.frame(
  SITE = c(1,2,3),
  Before=c(2,2,1),
  After=c(2,5,6),
  Yr_1=c(4,2,1),
  Yr_2=c(4,0,0),
  Yr_3=c(1,0,2),
  Yr_4=c(5,2,1),
  Yr_5=c(2,2,0),
  Yr_6=c(1,0,1),
  Yr_7=c(1,2,1))


df |> 
  select(-Before,-After) |> 
  pivot_longer(-SITE,names_prefix = 'Yr_',names_to = 'year') |> 
  group_by(SITE) |> 
  mutate(before = cumsum(value),
         before_no_current = cumsum(value) - value,
         after = sum(value) - before_no_current,
         after_no_current = after - value)
#> # A tibble: 21 x 7
#> # Groups:   SITE [3]
#>     SITE year  value before before_no_current after after_no_current
#>    <dbl> <chr> <dbl>  <dbl>             <dbl> <dbl>            <dbl>
#>  1     1 1         4      4                 0    18               14
#>  2     1 2         4      8                 4    14               10
#>  3     1 3         1      9                 8    10                9
#>  4     1 4         5     14                 9     9                4
#>  5     1 5         2     16                14     4                2
#>  6     1 6         1     17                16     2                1
#>  7     1 7         1     18                17     1                0
#>  8     2 1         2      2                 0     8                6
#>  9     2 2         0      2                 2     6                6
#> 10     2 3         0      2                 2     6                6
#> # ... with 11 more rows

^{由代表 package (v2.0.1) 於 2022 年 1 月 24 日創建}