R-时间变量和日期的条件运算 [英] R - Conditional operations on time variables and dates

问题描述

假设，我有下面的数据框，并且要对最后一个变量 HMS 做一些操作。
目的是创建一个新变量，假设 TimeStampNew 并指定第二天（如果观察值在 17:30:00之后） 。当观察值指示无值时（'NA'），则应使用同一日期的 TimeStamp 。

Assumed, I have the dataframe below and want to do some operations on the last variable HMS. The aim is to create a new variable, lets say TimeStampNew and assign the next day if the observation is after 17:30:00 of the day. When the observation indicates no value ('NA'), then the TimeStamp of the same date should be used.

问题是，要从类字符转换 HMS 变量，采用允许条件的格式。我已经检查了chron-package或将其转换为 POSIXct 类型的对象作为可能的选择。

The problem is, to convert the HMS-variable from the class character in a format allowing for conditions. I already checked the chron-package or converting it to a object of type POSIXct as possible alternatives.

我已经在这篇文章中问过类似的问题： R-字符变量 weekday的条件并以特殊的日期代替在数据框中。
我已经尝试通过 dplyr 包的某些操作来解决该问题，如上面提到的帖子中所建议。但是，由于时间格式复杂，我没有收到所需的解决方案。

A similar question I already asked in this post: R - Condition on character variable "weekday" and replace with special "date" in a data frame . I already tried to solve the issue with some operations of the dplyr package as suggested in the post mentioned above. However, I didn't receive the desired solution due to the tricky time format.

有人对这个问题有解决方案吗？！

Has anybody an solution proposal to this question?!

谢谢！

    TimeStamp      HMS
1  2010-02-22 19:55:00
2  2012-10-10 07:53:00
3  2012-10-24 07:55:00
4  2013-07-15 08:14:00
5  2013-07-24 08:23:00
6  2013-12-02 10:00:00
7  2012-02-23 07:56:00
8  2012-03-06 08:45:00
9  2013-09-15 19:54:00
10 2007-03-28 NA      
11 2007-08-09 NA      
12 2008-08-07 NA

推荐答案

使用 difftime 和 ifelse 语句：

df1$TimeStamp <- as.Date(df1$TimeStamp, format = "%Y-%m-%d")

df1$TimeStampNew <- ifelse(
  difftime(as.POSIXct(df1$HMS, format = "%H:%M:%S"), 
  as.POSIXct("17:30:00", format = "%H:%M:%S")) > 0,
  as.character(df1$TimeStamp + 1), 
  as.character(df1$TimeStamp))

df1$TimeStampNew[is.na(df1$HMS)] <- as.character(df1$TimeStamp[is.na(df1$HMS)])

它产生的数据框：

> df1
    TimeStamp      HMS TimeStampNew
1  2010-02-22 19:55:00   2010-02-23
2  2012-10-10 07:53:00   2012-10-10
3  2012-10-24 07:55:00   2012-10-24
4  2013-07-15 08:14:00   2013-07-15
5  2013-07-24 08:23:00   2013-07-24
6  2013-12-02 10:00:00   2013-12-02
7  2012-02-23 07:56:00   2012-02-23
8  2012-03-06 08:45:00   2012-03-06
9  2013-09-15 19:54:00   2013-09-16
10 2007-03-28     <NA>   2007-03-28
11 2007-08-09     <NA>   2007-08-09
12 2008-08-07     <NA>   2008-08-07

这篇关于R-时间变量和日期的条件运算的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！