如何筛选R数据框中每个国家最后一个非NA的Value值？

方法1：使用tidyverse（dplyr）包 #

这是日常数据分析中最常用的简洁写法：

# 首次使用时先安装包
# install.packages("dplyr")
library(dplyr)

# 原数据
df <- data.frame(  
        Country=c("Aruba","Aruba","Aruba","Afghanistan","Afghanistan","Afghanistan","Afghanistan","Butan","Butan","Belize"),
        Year=c("2007","2008","2009","2006","2007","2008","2009","2006","2009","2006"),
        Value=c(158,196,NA,156,140,693,854,NA,904,925))

# 筛选每个国家最后一个非NA的Value行
result <- df %>%
  # 将Year从字符型转为数值型，确保年份排序逻辑准确
  mutate(Year = as.numeric(Year)) %>%
  # 按国家分组
  group_by(Country) %>%
  # 过滤掉Value为NA的行
  filter(!is.na(Value)) %>%
  # 取每组的最后一行（即最新年份的有效数据）
  slice_tail(n = 1) %>%
  # 取消分组结构
  ungroup()

print(result)

执行后输出结果：

# A tibble: 4 × 3
  Country     Year Value
  <chr>      <dbl> <dbl>
1 Afghanistan  2009   854
2 Aruba        2008   196
3 Belize       2006   925
4 Butan        2009   904

方法2：基础R实现（无需额外包） #

如果不想依赖第三方包，用基础R函数也能完成：

# 原数据
df <- data.frame(  
        Country=c("Aruba","Aruba","Aruba","Afghanistan","Afghanistan","Afghanistan","Afghanistan","Butan","Butan","Belize"),
        Year=c("2007","2008","2009","2006","2007","2008","2009","2006","2009","2006"),
        Value=c(158,196,NA,156,140,693,854,NA,904,925))

# 转换Year为数值型
df$Year <- as.numeric(df$Year)

# 按国家分组处理，过滤非NA行后取最后一行
result <- do.call(rbind, lapply(split(df, df$Country), function(group) {
  non_na_rows <- group[!is.na(group$Value), ]
  # 避免无有效数据的国家报错
  if (nrow(non_na_rows) > 0) tail(non_na_rows, 1) else NULL
}))

print(result)

方法3：data.table包（适合大数据集） #

如果处理百万级以上的大数据，data.table的运行效率更高：

# 首次使用时先安装包
# install.packages("data.table")
library(data.table)

# 原数据
df <- data.frame(  
        Country=c("Aruba","Aruba","Aruba","Afghanistan","Afghanistan","Afghanistan","Afghanistan","Butan","Butan","Belize"),
        Year=c("2007","2008","2009","2006","2007","2008","2009","2006","2009","2006"),
        Value=c(158,196,NA,156,140,693,854,NA,904,925))

# 转换为data.table对象并处理Year列
setDT(df)[, Year := as.numeric(Year)]

# 按国家分组，过滤非NA后取每组最后一行
result <- df[!is.na(Value), .SD[.N], by = Country]

print(result)

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