Question

在尝试按月汇总数据时，我在处理R中的这一数据管理步骤时遇到了困难。

我有两个例子，一个基本上代表我现在在清理和聚合过程中遇到的问题，第二个代表我希望它看起来像什么。

现在的样子：

month <- c("January", "January", "February", "March", "April", "April", 
"May", "June", "July")
year <- c(2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014)
count1 <- c(3, 0, 1, 2, 0, 8, 1, 1, 1)
count2 <- c(0, 2, 1, 4, 6, 0, 1, 1, 1)
count3 <- c(1, 1, 1, 1, 1, 1, 0, 0, 1)

df <- data.frame(month, year, count1, count2, count3)

我希望它看起来像：

month2 <- c("January", "February", "March", "April", "May", "June", "July")
year2 <- c(2014, 2014, 2014, 2014, 2014, 2014, 2014)
count1a <- c(3, 1, 2, 8, 1, 1, 1)
count2a <- c(2, 1, 4, 6, 1, 1, 1)
count3a <- c(1, 1, 1, 1, 0, 0, 1)

df2 <- data.frame(month2, year2, count1a, count2a, count3a)

正如您将注意到的那样，我有几个月被计算两次，他们的观察与我在同一个月使用的其他观察结果不同。

Answer 1

按“月”和“年”分组后，获取max

library(dplyr)
df %>% 
   group_by(month, year) %>%
   summarise_all(max)
# A tibble: 7 x 5
# Groups:   month [?]
#     month  year count1 count2 count3
#    <fctr> <dbl>  <dbl>  <dbl>  <dbl>
#1    April  2014      8      6      1
#2 February  2014      1      1      1
#3  January  2014      3      2      1
#4     July  2014      1      1      1
#5     June  2014      1      1      0
#6    March  2014      2      4      1
#7      May  2014      1      1      0

如果我们需要保持相同的顺序，那么

df %>% 
   group_by(month = factor(month, levels = unique(month)), year) %>% 
   summarise_all(max)
   #or
   #summarise_all(funs(.[order(-.)][1]))

R中的数据管理 - 聚合后整理数据

1 个答案: