R Group By取第一条记录

id = c(1,2,1,4,2)
year = c(1980, 1990, 2000, 2010, 2020)
salary = c(2000, 3000, 4000, 5000, 6000)
df = data.frame(id, year, salary);
df
# id year salary
# 1  1 1980   2000
# 2  2 1990   3000
# 3  1 2000   4000
# 4  4 2010   5000
# 5  2 2020   6000

df %>%  group_by(id) %>%
  arrange(id, desc(year)) %>%
  filter(row_number() == 1) %>%
  ungroup()

# id  year salary
# <dbl> <dbl>  <dbl>
#   1  2000   4000
#   2  2020   6000
#   4  2010   5000

df %>%  group_by(id) %>%  #按id分组
  arrange(id, desc(year)) %>% #按id升序，year降序
  filter(row_number() == 1) %>% #取分组后的第一条记录，即日期最近的一条记录
  ungroup() #取消分组