计算R中多个变化列之间的差异

你可以试试 ifelse

transform(
  df,
  Difference = ifelse(
    is.na(Year3),
    (Size2 - Size1) / (Year2 - Year1),
    (Size3 - Size1) / (Year3 - Year1)
  )
)

这给了

  ID Year1 Year2 Year3 Size1 Size2 Size3 Difference
1  A  2021  2022    NA    10    15    NA        5.0
2  B  2022  2023  2024    20    21    25        2.5
3  C  2021  2022    NA     5    20    NA       15.0

如果,如问题中所述,以下两个或两个都不是真的 Year3 和 Size3 如果是NA,那么我们可以使用 coalesce . 聚结 如果它不是NA,则返回其第一个参数,否则返回其第二个参数。输入在末尾的注释中提供。

library(dplyr)

# check that both or neither Year3 and Size3 are NA
with(df, all(is.na(Year3) == is.na(Size3)))
## [1] TRUE

df %>%
  mutate(Difference = (coalesce(Size3, Size2) - Size1) / 
    (coalesce(Year3, Year2) - Year1))
##   ID Year1 Year2 Year3 Size1 Size2 Size3 Difference
## 1  A  2021  2022    NA    10    15    NA        5.0
## 2  B  2022  2023  2024    20    21    25        2.5
## 3  C  2021  2022    NA     5    20    NA       15.0

注:

可复制形式的输入:

df <- data.frame(
  ID = c("A", "B", "C"),
  Year1 = c(2021L, 2022L, 2021L),
  Year2 = c(2022L, 2023L, 2022L),
  Year3 = c(NA, 2024L, NA),
  Size1 = c(10L, 20L, 5L),
  Size2 = c(15L, 21L, 20L),
  Size3 = c(NA, 25L, NA)
)