R语言mutate+case_when报错“Undefined columns selected”求助

执行以下代码时出现错误：

test %>% mutate(Value = case_when(row_number() == 1 ~ val_1,
                                      row_number() == 2 ~ val_2))

错误信息：

Error in `mutate()`:
! Problem while computing `Value = case_when(row_number() == 1 ~ val_1, row_number() == 2 ~ 
val_2)`.
Caused by error in `[.data.frame`:
! undefined columns selected

test数据框结构：

structure(list(ID = c("47ae3cc00d8515d4cbab4704bf7e6b79", "47ae3cc00d8515d4cbab4704bf7e6b79", 
"47ae3cc00d8515d4cbab4704bf7e6b79", "47ae3cc00d8515d4cbab4704bf7e6b79"
), Book = c("MyBookie.ag", "MyBookie.ag", "Pinnacle", "Pinnacle"
), Home = c("Kansas St Wildcats", "Kansas St Wildcats", "Kansas St Wildcats", 
"Kansas St Wildcats"), Away = c("Michigan St Spartans", "Michigan St Spartans", 
"Michigan St Spartans", "Michigan St Spartans"), Team = c("Kansas St Wildcats", 
"Michigan St Spartans", "Kansas St Wildcats", "Michigan St Spartans"
), Price = c(-110, -110, -114, 100), Points = c(1.5, -1.5, 1.5, 
-1.5), BEP = c(0.523809524, 0.523809524, 0.53271028, 0.5), Fair_Price = c(100, 
100, -106.5420561, 106.5420561), Fair_BEP = c(0.5, 0.5, 0.515837104, 
0.484162896), Hold = c(0.045454545, 0.045454545, 0.031674208, 
0.031674208), Win = c(90.91, 90.91, 87.72, 100)), row.names = c(NA, 
-4L), spec = structure(list(cols = list(ID = structure(list(), class = 
c("collector_character", 
"collector")), Book = structure(list(), class = c("collector_character", 
"collector")), Home = structure(list(), class = c("collector_character", 
"collector")), Away = structure(list(), class = c("collector_character", 
"collector")), Team = structure(list(), class = c("collector_character", 
"collector")), Price = structure(list(), class = c("collector_double", 
"collector")), Points = structure(list(), class = c("collector_double", 
"collector")), BEP = structure(list(), class = c("collector_double", 
"collector")), Fair_Price = structure(list(), class = c("collector_double", 
"collector")), Fair_BEP = structure(list(), class = c("collector_double", 
"collector")), Hold = structure(list(), class = c("collector_double", 
"collector")), Win = structure(list(), class = c("collector_double", 
"collector"))), default = structure(list(), class = c("collector_guess", 
"collector")), delim = ","), class = "col_spec"), problems = <pointer: 0x00000153a4920d20>, 
class = c("spec_tbl_df", 
"tbl_df", "tbl", "data.frame"))

val_1和val_2的定义及结构：

# val_1定义与结构
val_1 <- (test[4,10] * test[2,11]) - (test[3,10] * 100)
structure(list(Fair_BEP = -51.5617029958564), class = "data.frame", row.names = c(NA, 
-1L))

# val_2定义与结构
val_2 <- (test[3,10] * test[1,11]) - (test[4,10] * 100)
structure(list(Fair_BEP = -48.3928424591436), class = "data.frame", row.names = c(NA, 
-1L))

问题核心：val_1和val_2是单行列的数据框，而case_when要求右侧必须是标量或长度匹配的向量，二维结构的data.frame无法直接参与赋值逻辑，导致报错。

方法1：提取标量值重新定义val变量 #

修改val的定义，用[[或pull()提取单个数值（标量）：

# 用[[直接提取标量
val_1 <- (test[[4,10]] * test[[2,11]]) - (test[[3,10]] * 100)
val_2 <- (test[[3,10]] * test[[1,11]]) - (test[[4,10]] * 100)

# 或用dplyr的pull函数，语义更清晰
val_1 <- (test %>% slice(4) %>% pull(Fair_BEP)) * (test %>% slice(2) %>% pull(Hold)) - (test %>% slice(3) %>% pull(Fair_BEP)) * 100
val_2 <- (test %>% slice(3) %>% pull(Fair_BEP)) * (test %>% slice(1) %>% pull(Hold)) - (test %>% slice(4) %>% pull(Fair_BEP)) * 100

# 执行原mutate代码即可正常运行
test %>% mutate(Value = case_when(row_number() == 1 ~ val_1,
                                  row_number() == 2 ~ val_2,
                                  TRUE ~ NA_real_))

方法2：在mutate内部直接计算，避免外部变量 #

利用dplyr的nth()函数，在当前数据框上下文直接提取指定行的数值，无需提前定义val变量：

test %>% 
  mutate(
    Value = case_when(
      row_number() == 1 ~ (nth(Fair_BEP, 4) * nth(Hold, 2)) - (nth(Fair_BEP, 3) * 100),
      row_number() == 2 ~ (nth(Fair_BEP, 3) * nth(Hold, 1)) - (nth(Fair_BEP, 4) * 100),
      TRUE ~ NA_real_ # 为其余行设置NA，避免长度不匹配报错
    )
  )

方法3：直接构造赋值向量 #

因为仅需给前两行赋值，直接构造对应长度的向量完成赋值：

test %>% 
  mutate(Value = c(
    (test[[4,10]] * test[[2,11]]) - (test[[3,10]] * 100),
    (test[[3,10]] * test[[1,11]]) - (test[[4,10]] * 100),
    NA_real_, NA_real_
  ))

内容的提问来源于stack exchange，提问作者Aaron Morris