根据 R 中的值和 NA 过滤 "POSIXct" "POSIXt" 列

Filtering "POSIXct" "POSIXt" column based on value and NA in R

我有一个大致像这样的数据框:

    tail(df)
    # A tibble: 6 x 3
      GEOGCD    OPER_DATE           TERM_DATE          
      <chr>     <dttm>              <dttm>             
    1 E05006867 2009-01-01 00:00:00 2019-03-31 00:00:00
    2 E05006868 2009-01-01 00:00:00 2019-03-31 00:00:00
    3 E05000066 2009-01-01 00:00:00 2018-05-02 00:00:00
    4 E05000067 2009-01-01 00:00:00 2018-05-02 00:00:00
    5 E05000068 2009-01-01 00:00:00 2018-05-02 00:00:00
    6 E05000064 2018-05-01 22:00:00 NA          
    
    str(df)
    tibble [52 × 3] (S3: tbl_df/tbl/data.frame)
     $ GEOGCD   : chr [1:52] "E05000064" "E05000065" "E05000066" "E05000067" ...
     $ OPER_DATE: POSIXct[1:52], format: "2009-01-01 00:00:00" "2009-01-01 00:00:00" "2009-01-01 00:00:00" ...
     $ TERM_DATE: POSIXct[1:52], format: "2018-05-02" "2018-05-02" "2018-05-02" ...

我想做的是 select 只有 TERM_DATE 高于 2018-12-31 或者是 NA 的人。基本上是这样的:

3 E05000066 2009-01-01 00:00:00 2018-05-02 00:00:00
4 E05000067 2009-01-01 00:00:00 2018-05-02 00:00:00
5 E05000068 2009-01-01 00:00:00 2018-05-02 00:00:00
6 E05000064 2018-05-01 22:00:00 NA   

我尝试过不同的方法,例如:

library(lubridate)
library(dplyr)    
df%>%
filter(TERM_DATE> as.Date("2018-12-31"| is.na(TERM_DATE)))

但我不断收到如下错误:

Error: Problem with filter() input ..1.
x operations are possible only for numeric, logical or complex types
ℹ Input ..1 is TERM_DATE > as.Date("2018-12-31" | is.na(TERM_DATE)).

你们中有人能理解为什么会这样吗?我应该怎么做?

谢谢!

试试这个方法:

library(dplyr)
#Code
newdf <- df%>%
  filter(TERM_DATE> as.POSIXct("2018-12-31") | is.na(TERM_DATE))

输出:

     GEOGCD           OPER_DATE  TERM_DATE
1 E05006867 2009-01-01 00:00:00 2019-03-31
2 E05006868 2009-01-01 00:00:00 2019-03-31
3 E05000064 2018-05-01 22:00:00       <NA>

来自 @StupidWolf 的智能解决方案也有效:

#Code 2
df%>%
  filter(TERM_DATE> as.Date("2018-12-31") | is.na(TERM_DATE))

输出:

     GEOGCD           OPER_DATE  TERM_DATE
1 E05006867 2009-01-01 00:00:00 2019-03-31
2 E05006868 2009-01-01 00:00:00 2019-03-31
3 E05000064 2018-05-01 22:00:00       <NA>

OP 的预期输出可以使用:

#Code 3
newdf <- df%>%
  filter(TERM_DATE< as.POSIXct("2018-12-31") | is.na(TERM_DATE))

输出:

     GEOGCD           OPER_DATE  TERM_DATE
1 E05000066 2009-01-01 00:00:00 2018-05-02
2 E05000067 2009-01-01 00:00:00 2018-05-02
3 E05000068 2009-01-01 00:00:00 2018-05-02
4 E05000064 2018-05-01 22:00:00       <NA>

或使用as.Date()。您需要将比较更改为 <.