使用 R 中的 2 个条件匹配表答案

【问题标题】：Match tables using 2 criteria in R使用 R 中的 2 个条件匹配表
【发布时间】：2020-04-01 16:44:16
【问题描述】：

我刚开始在 R 中编码，我正在尝试操作数据，但我遇到了以下问题：我有 2 个不同的表（简化）第一个（player_df）如下：

 name   experience      Club        age       Position 
 luc          2         FCB         18        Goalkeeper
 jean         9         Real        26        midfielder
 ronaldo      14        FCB         32        Goalkeeper
 jean         9         Real        26        midfielder
 messi        11        Liverpool   35        midfielder
 tevez        6         Chelsea     27        Attack
 inzaghi      9         Juve        34        Defender
 kwfni        17        Bayern      40        Attack
 Blabla       9         Real       25        midfielder
 wdfood      11        Liverpool   33        midfielder
 player2      7         Chelsea     28       Attack
 player3     10         Juve       34        Defender
 fgh         17        Bayern      40        Attack
...

第二张表是按俱乐部和经验分的薪水，单位是百万（salary_df）

*experience    FCB   BAYERN    Juve   Real  Chelsea 
1               1.5   1.3     1      4      3
2               2.5   2       2.4    5      4
3               3.4   3.1     3.5    6.3    5
4               5     4.5     6.7     9     6
5               7.1   6.9     9      12     7
6               9      8      10     15     10
7               10     9      12     16     15
8               14     12     13     19     16
9               14.5   17     15     20     17
10              15     19     17     23     18
..*

我想在名为 let say salary_estimation 的第一个表中为我的数据添加一个新列，其中考虑了 2 个变量，例如 experience 和 club。

例如对于在“FCB”打球并有“2”年经验的“luc”，输出应该是“2.5”

在 excel 中它是一个索引/匹配函数，但在 R 中我不知道应该使用哪个函数。

我应该如何解决这个问题？

【问题讨论】：

标签： r join merge

【解决方案1】：

数据：

df1 <- read.table(text = 'name   experience      Club        age       Position 
 luc          2         FCB         18        Goalkeeper
                 jean         9         Real        26        midfielder
                 ronaldo      14        FCB         32        Goalkeeper
                 jean         9         Real        26        midfielder
                 messi        11        Liverpool   35        midfielder
                 tevez        6         Chelsea     27        Attack
                 inzaghi      9         Juve        34        Defender
                 kwfni        17        Bayern      40        Attack
                 Blabla       9         Real       25        midfielder
                 wdfood      11        Liverpool   33        midfielder
                 player2      7         Chelsea     28       Attack
                 player3     10         Juve       34        Defender
                 fgh         17        Bayern      40        Attack', header = TRUE, stringsAsFactors = FALSE)

df2 <- read.table(text = 'experience    FCB   BAYERN    Juve   Real  Chelsea 
1               1.5   1.3     1      4      3
                  2               2.5   2       2.4    5      4
                  3               3.4   3.1     3.5    6.3    5
                  4               5     4.5     6.7     9     6
                  5               7.1   6.9     9      12     7
                  6               9      8      10     15     10
                  7               10     9      12     16     15
                  8               14     12     13     19     16
                  9               14.5   17     15     20     17
                  10              15     19     17     23     18', header = TRUE, stringsAsFactors = FALSE)

代码：

library('data.table')
setDT(df2)[, Chelsea := as.numeric(Chelsea)]
df2 <- melt(df2, id.vars = "experience", variable.name = "Club", value.name = "Salary" )
df2[df1, on = c("experience", "Club"), nomatch = NA]

输出：

#    experience      Club Salary    name age   Position
# 1:          2       FCB    2.5     luc  18 Goalkeeper
# 2:          9      Real   20.0    jean  26 midfielder
# 3:         14       FCB     NA ronaldo  32 Goalkeeper
# 4:          9      Real   20.0    jean  26 midfielder
# 5:         11 Liverpool     NA   messi  35 midfielder
# 6:          6   Chelsea   10.0   tevez  27     Attack
# 7:          9      Juve   15.0 inzaghi  34   Defender
# 8:         17    Bayern     NA   kwfni  40     Attack
# 9:          9      Real   20.0  Blabla  25 midfielder
# 10:         11 Liverpool     NA  wdfood  33 midfielder
# 11:          7   Chelsea   15.0 player2  28     Attack
# 12:         10      Juve   17.0 player3  34   Defender
# 13:         17    Bayern     NA     fgh  40     Attack

【讨论】：

【解决方案2】：

一种可能的解决方案是使用experience 和club 作为键，将第一个表（假设它是player_df）与第二个表salary_df 的“长格式”连接起来。您可以使用tidyverse 包来实现。

library(tidyverse)

player_df %>%
  mutate(Club = str_to_title(Club)) %>%
  left_join(
      salary_df %>% 
        pivot_longer(-experience, names_to = "Club", values_to = "salary_estimation") %>%
        mutate(Club = str_to_title(Club)) )

# Joining, by = c("experience", "Club")
# # A tibble: 13 x 6
#    name    experience Club        age Position   salary_estimation
#    <chr>        <dbl> <chr>     <dbl> <chr>                  <dbl>
#  1 luc              2 Fcb          18 Goalkeeper               2.5
#  2 jean             9 Real         26 midfielder              20  
#  3 ronaldo         14 Fcb          32 Goalkeeper              NA  
#  4 jean             9 Real         26 midfielder              20  
#  5 messi           11 Liverpool    35 midfielder              NA  
#  6 tevez            6 Chelsea      27 Attack                  10  
#  7 inzaghi          9 Juve         34 Defender                15  
#  8 kwfni           17 Bayern       40 Attack                  NA  
#  9 Blabla           9 Real         25 midfielder              20  
# 10 wdfood          11 Liverpool    33 midfielder              NA  
# 11 player2          7 Chelsea      28 Attack                  15  
# 12 player3         10 Juve         34 Defender                17  
# 13 fgh             17 Bayern       40 Attack                  NA

【讨论】：