在数据框中查找每个月的最后一个日期

我一直在工作中遇到一个问题。我有一个日期数据集（采用数据框格式），我需要搜索并找到每个月的最后一天并将它们放入新的数据框中。下一列中还有一个与之相关的值。这是日期的样本。

   [1] "2015-05-21" "2015-05-20" "2015-05-19" "2015-05-18" "2015-05-15" "2015-05-14"
   [7] "2015-05-13" "2015-05-12" "2015-05-11" "2015-05-08" "2015-05-07" "2015-05-06"
  [13] "2015-05-05" "2015-05-04" "2015-05-01" "2015-04-30" "2015-04-29" "2015-04-28"
  [19] "2015-04-27" "2015-04-24" "2015-04-23" "2015-04-22" "2015-04-21" "2015-04-20"
  [25] "2015-04-17" "2015-04-16" "2015-04-15" "2015-04-14" "2015-04-13" "2015-04-10"
  [31] "2015-04-09" "2015-04-08" "2015-04-07" "2015-04-06" "2015-04-02" "2015-04-01"
  [37] "2015-03-31" "2015-03-30" "2015-03-27" "2015-03-26" "2015-03-25" "2015-03-24"
  [43] "2015-03-23" "2015-03-20" "2015-03-19" "2015-03-18" "2015-03-17" "2015-03-16"
  [49] "2015-03-13" "2015-03-12" "2015-03-11" "2015-03-10" "2015-03-09" "2015-03-06"
  [55] "2015-03-05" "2015-03-04" "2015-03-03" "2015-03-02" "2015-02-27" "2015-02-26"
  [61] "2015-02-25" "2015-02-24" "2015-02-23" "2015-02-20" "2015-02-19" "2015-02-18"
  [67] "2015-02-17" "2015-02-13" "2015-02-12" "2015-02-11" "2015-02-10" "2015-02-09"
  [73] "2015-02-06" "2015-02-05" "2015-02-04" "2015-02-03" "2015-02-02" "2015-01-30"
  [79] "2015-01-29" "2015-01-28" "2015-01-27" "2015-01-26" "2015-01-23" "2015-01-22"
  [85] "2015-01-21" "2015-01-20" "2015-01-16" "2015-01-15" "2015-01-14" "2015-01-13"
  [91] "2015-01-12" "2015-01-09" "2015-01-08" "2015-01-07" "2015-01-06" "2015-01-05"
  [97] "2015-01-02" "2014-12-31" "2014-12-30" "2014-12-29" "2014-12-26" "2014-12-24"
 [103] "2014-12-23" "2014-12-22" "2014-12-19" "2014-12-18" "2014-12-17" "2014-12-16"
 [109] "2014-12-15" "2014-12-12" "2014-12-11" "2014-12-10" "2014-12-09" "2014-12-08"
 [115] "2014-12-05" "2014-12-04" "2014-12-03" "2014-12-02" "2014-12-01" "2014-11-28"
 [121] "2014-11-26" "2014-11-25" "2014-11-24" "2014-11-21" "2014-11-20" "2014-11-19"
 [127] "2014-11-18" "2014-11-17" "2014-11-14" "2014-11-13" "2014-11-12" "2014-11-11"
 [133] "2014-11-10" "2014-11-07" "2014-11-06" "2014-11-05" "2014-11-04" "2014-11-03"
 [139] "2014-10-31" "2014-10-30" "2014-10-29" "2014-10-28" "2014-10-27" "2014-10-24"
 [145] "2014-10-23" "2014-10-22" "2014-10-21" "2014-10-20" "2014-10-17" "2014-10-16"
 [151] "2014-10-15" "2014-10-14" "2014-10-13" "2014-10-10" "2014-10-09" "2014-10-08"
 [157] "2014-10-07" "2014-10-06" "2014-10-03" "2014-10-02" "2014-10-01" "2014-09-30"
 [163] "2014-09-29" "2014-09-26" "2014-09-25" "2014-09-24" "2014-09-23" "2014-09-22"
 [169] "2014-09-19" "2014-09-18" "2014-09-17" "2014-09-16" "2014-09-15" "2014-09-12"

这是一小部分。有5700行...

1) Try tapply像这样：

dates <- c("2015-05-13", "2015-05-12", "2015-05-11", "2015-04-27", 
      "2015-04-24", "2015-04-23")

tapply(dates, substr(dates, 1, 7), max)

or this:

library(zoo)
tapply(dates, as.yearmon(dates), max)

2)如果日期位于数据框中：

 DF <- data.frame(dates, stringsAsFactors = FALSE)

 aggregate(DF["dates"], list(month = substr(DF$dates, 1, 7)), max)

或更换substr(...) with as.yearmon(DF$dates).

3)这会选出每个月的最后一行。在这种情况下ave返回一个字符变量所以需要使用as.logical将其转换为逻辑：

isMax <- function(x) seq_along(x) == which.max(as.Date(x))
subset(DF, as.logical(ave(dates, substr(dates, 1, 7), FUN = isMax)))

or use as.yearmon(dates)代替substr(...)。以下定义isMax可以替代

isMax <- function(x) seq_along(x) == which.max(as.Date(x))

并给出相同的结果，除非有多个最大值。在这种情况下，它仅给出第一个，而早期的定义给出了全部。作为which.max不适用于"character"我们将日期转换为的类"Date" class.

下次请使用dput显示您的数据或以可复制的形式提供数据，就像我们上面所做的那样。还将其减少到说明问题所需的最小数量。