Question

我有一个这样的数据框：

dfin <- structure(list(stock = structure(1:3, .Label = c("stock1", "stock2", 
"stock3"), class = "factor"), nameA = c(0.2, 0.3, 1.4), another = c(2L, 
1L, 3L), thirdname = c(0L, 4L, 3L)), .Names = c("stock", "nameA", 
"another", "thirdname"), class = "data.frame", row.names = c(NA, 
-3L))
#> dfin
#   stock nameA another thirdname
#1 stock1   0.2       2         0
#2 stock2   0.3       1         4
#3 stock3   1.4       3         3

我想计算每一行，将库存作为索引列，使用其他值的变量/列来计算这个类型：

typeA = nameA - another + thirdname
second = -nameA + another - thirdname

我想生成这样的输出数据框：

dfout <- structure(list(stock = structure(1:3, .Label = c("stock1", "stock2", 
"stock3"), class = "factor"), typeA = structure(c(1L, 3L, 2L), .Label = c("-1,8", 
"1,4", "3,3"), class = "factor"), second = structure(c(3L, 2L, 
1L), .Label = c("-1,4", "-3,3", "1,8"), class = "factor")), .Names = c("stock", 
"typeA", "second"), class = "data.frame", row.names = c(NA, -3L
))
> dfout
   stock typeA second
1 stock1  -1,8    1,8
2 stock2   3,3   -3,3
3 stock3   1,4   -1,4

有没有简单的方法可以为数据帧的所有行创建它而无需使用for循环？

Answer 1

使用基数R，您可以：

within(dfin, {
  typeA = nameA - another + thirdname
  second = -nameA + another - thirdname
  rm(nameA, another, thirdname)
})
#    stock second typeA
# 1 stock1    1.8  -1.8
# 2 stock2   -3.3   3.3
# 3 stock3   -1.4   1.4

使用tidyverse，你可以这样做：

library(tidyverse)

dfin %>%
  mutate(typeA = nameA - another + thirdname,
         second = -nameA + another - thirdname) %>%
  select(-nameA, -another, -thirdname)

或根据评论中的建议transmute：

dfin %>%
  transmute(stock,
            typeA = nameA - another + thirdname,
            second = -nameA + another - thirdname)

使用“data.table”，例如：

library(data.table)
as.data.table(dfin)[, c("typeA", "second") := list(
  nameA - another + thirdname,
  -nameA + another - thirdname
)][, c("nameA", "another", "thirdname") := NULL][]

计算数据帧的每一行的方程

1 个答案: