Time and date

Create and format date values

Class Date

Sys.Date()

[1] "2020-12-08"

(myDate <- as.Date("01.11.1974", format="%d.%m.%Y"))

[1] "1974-11-01"

format(myDate, format="%d.%m.%Y")

[1] "01.11.1974"

Internal representation

(negDate <- as.Date(-374, "1910-12-16"))

[1] "1909-12-07"

as.numeric(negDate)

[1] -21940

Time values

Class POSIXct

Sys.time()

[1] "2020-12-08 16:32:22 CET"

date()

[1] "Tue Dec  8 16:32:22 2020"

(myTime <- as.POSIXct("2009-02-07 09:23:02"))

[1] "2009-02-07 09:23:02 CET"

ISOdate(2010, 6, 30, 17, 32, 10, tz="CET")

[1] "2010-06-30 17:32:10 CEST"

format(myTime, "%H:%M:%S")

[1] "09:23:02"

format(myTime, "%d.%m.%Y")

[1] "07.02.2009"

Class POSIXlt

charDates <- c("05.08.1972, 03:37", "02.04.1981, 12:44")
(lDates   <- strptime(charDates, format="%d.%m.%Y, %H:%M"))

[1] "1972-08-05 03:37:00 CET"  "1981-04-02 12:44:00 CEST"

lDates$mday

[1] 5 2

lDates$hour

[1]  3 12

weekdays(lDates)

[1] "Saturday" "Thursday"

months(lDates)

[1] "August" "April" 

Time and date arithmetic

Sum and difference of time-date values

(myDate <- as.Date("01.11.1974", format="%d.%m.%Y"))

[1] "1974-11-01"

myDate + 365

[1] "1975-11-01"

(diffDate <- as.Date("1976-06-19") - myDate)

Time difference of 596 days

as.numeric(diffDate)

[1] 596

myDate + diffDate

[1] "1976-06-19"

lDates + c(60, 120)

[1] "1972-08-05 03:38:00 CET"  "1981-04-02 12:46:00 CEST"

(diff21 <- lDates[2] - lDates[1])

Time difference of 3162.338 days

lDates[1] + diff21

[1] "1981-04-02 12:44:00 CEST"

round(lDates, units="days")

[1] "1972-08-05 CET"  "1981-04-03 CEST"

trunc(lDates, units="years")

[1] "1972-01-01 CET" "1981-01-01 CET"

Systematically and randomly generate time-date values

seq(ISOdate(2010, 5, 1), ISOdate(2013, 5, 1), by="years")

[1] "2010-05-01 12:00:00 GMT" "2011-05-01 12:00:00 GMT"
[3] "2012-05-01 12:00:00 GMT" "2013-05-01 12:00:00 GMT"

seq(ISOdate(1997, 10, 22), by="2 weeks", length.out=4)

[1] "1997-10-22 12:00:00 GMT" "1997-11-05 12:00:00 GMT"
[3] "1997-11-19 12:00:00 GMT" "1997-12-03 12:00:00 GMT"

secsPerDay <- 60 * 60 * 24
randDates  <- ISOdate(1995, 6, 13) + sample(0:(28*secsPerDay), 100, replace=TRUE)
randWeeks  <- cut(randDates, breaks="week")
summary(randWeeks)

1995-06-12 1995-06-19 1995-06-26 1995-07-03 1995-07-10 
        19         25         29         24          3 

Useful packages

Package lubridate provides many functions for efficiently and consistently handling times and dates. More packages can be found in CRAN task view Time Series.

Get the article source from GitHub

R markdown - markdown - R code - all posts