CODEWITHSUNDEEP
rtimeposixct
ay__ya
ay__ya

Reputation: 463

Distinguish Peak and Off-Peak traffic periods referring to POSIXct format time

Here I have some air pollution concentration data with number concentration per minute collected in London. This pollutant is a traffic-related air pollutant. I would like to add another column with a factor variable (Peak; Off-Peak) to distinguish the data was collected during Peak time (6:30 am-9:30 am) or Off-Peak time (16:00 to 19:00) only for Monday to Friday. The current time column is in POSIXct format with date+time.

Sample Data

QUESTION 1: Should I do the following: 1.remove Saturday and Sunday data 2.extract time from the date+time 3.identify time range of (6:30 am-9:30 am) and (16:00 to 19:00)

OR there is a way to identify time range of a day using date+time (POSIXct)

QUESTION 2:How can I properly extract time from date+time variable of POSIXct format, which can be used for time range ((6:30 am-9:30 am) and (16:00 to 19:00)) identification?

Upvotes: 0

Views: 65

Answers (1)

Edward
Edward

Reputation: 18708

You can use the chron package as it has a function for determining weekdays. It also has a function for holidays. And the lubridate package can be used to extract the hour and minute from the date/time variable.

library(chron)
library(lubridate)

data %>%
  mutate(peak = case_when(
    is.weekend(as.Date(date)) ~ FALSE,
    (hour(date)==6 & minute(date)>=30 | hour(date)>6) & 
      (hour(date)==9 & minute(date)<=30 | hour(date)<9) ~ TRUE,
    TRUE ~ FALSE),
    offpeak = case_when(
      is.weekend(as.Date(date)) ~ FALSE,
      (hour(date)>=16 & hour(date)<=19) ~ TRUE,
      TRUE ~ FALSE)
    )

                  date  peak offpeak
1  1998-01-01 06:00:00 FALSE   FALSE
2  1998-01-01 06:20:00 FALSE   FALSE
3  1998-01-01 06:40:00  TRUE   FALSE
4  1998-01-01 07:00:00  TRUE   FALSE
5  1998-01-01 07:20:00  TRUE   FALSE
6  1998-01-01 07:40:00  TRUE   FALSE
7  1998-01-01 08:00:00  TRUE   FALSE
8  1998-01-01 08:20:00  TRUE   FALSE
9  1998-01-01 08:40:00  TRUE   FALSE
10 1998-01-01 09:00:00  TRUE   FALSE
11 1998-01-01 09:20:00  TRUE   FALSE
12 1998-01-01 09:40:00 FALSE   FALSE
13 1998-01-01 10:00:00 FALSE   FALSE

30 1998-01-01 15:40:00 FALSE   FALSE
31 1998-01-01 16:00:00 FALSE    TRUE
32 1998-01-01 16:20:00 FALSE    TRUE
33 1998-01-01 16:40:00 FALSE    TRUE
34 1998-01-01 17:00:00 FALSE    TRUE
35 1998-01-01 17:20:00 FALSE    TRUE
36 1998-01-01 17:40:00 FALSE    TRUE
37 1998-01-01 18:00:00 FALSE    TRUE
38 1998-01-01 18:20:00 FALSE    TRUE
39 1998-01-01 18:40:00 FALSE    TRUE
40 1998-01-01 19:00:00 FALSE    TRUE
41 1998-01-01 19:20:00 FALSE    TRUE
42 1998-01-01 19:40:00 FALSE    TRUE
43 1998-01-01 20:00:00 FALSE   FALSE

Data:

data <- data.frame(date=seq.POSIXt(as.POSIXct("1998-01-01 06:00:00"), 
                                   as.POSIXct("1998-01-02 06:00:00"), by="20 min"))

Upvotes: 0

Related Questions