Get started with datey

library(datey)

This vignette is a hands-on guide to the datey package. For the motivation behind the annual-grid approach and the associativity guarantee, see Why datey?. For the complete formal specification, see the datey specification.

The core types

datey provides three S3 classes:

datey – a point in time, stored at day-fraction precision.
durationy – a duration in years.
datey_interval – a half-open [start, end) time interval.

These are atomic types¹ that store dates and durations as integers with units of 1/534 360 of a year). As a result, arithmetic with these types is exact and associative.

Creating a `datey`

When in the day?

Exposure periods specified by the dates to typically mean that the whole of the day and the whole of the day are included. In the datey system this corresponds to using start_day() for and end_day() for .

Deaths on the other hand typically happen during a day. In the datey system this corresponds to using mid_day().

These distinctions may be new to you and your first reaction may be that they are immaterial. But it costs very little to be precise and sometimes systematic errors can accumulate and end up being material.

From year, month and day

start_day(), mid_day() and end_day() create a datey from scratch:

start_day(2024, 3, 7)   # Start of the day 7 March 2024
#> [1] 2024-03-07.0
mid_day(2024, 3, 7)     # Middle of the day 7 March 2024
#> [1] 2024-03-07.5
end_day(2024, 3, 7)     # End of the day 7 March 2024
#> [1] 2024-03-08.0

The end of a day is the same point as the start of the next, so end_day() applied to a day is identical to start_day() applied to the following day:

identical(end_day(2024, 3, 7), start_day(2024, 3, 8))
#> [1] TRUE

For an arbitrary position within a day, datey() accepts a day fraction between 0 and 1:

datey(2024, 3, 7, 0.25)   # Quarter of the way through the day 7 March 2024
#> [1] 2024-03-07.25

From base R dates

It is often the case that data already contains dates defined using the standard base R types Date², POSIXct or POSIXlt.

To convert these to a datey, use start_day(), mid_day() or end_day():

d <- as.Date("2024-03-07")
start_day(d)
#> [1] 2024-03-07.0
mid_day(d)
#> [1] 2024-03-07.5
end_day(d)
#> [1] 2024-03-08.0

From fractional years or text

datey() also accepts a fractional calendar year or a character string in YYYY-MM-DD[.f] format:

datey(2024)           # Start of calendar year 2024
#> [1] 2024-01-01.0
datey(2024.5)         # Midway through calendar year 2024
#> [1] 2024-07-02.0

datey("2024-03-07")   # Start of the day 7 March 2024 (day fraction defaults to 0)
#> [1] 2024-03-07.0
datey("2024-03-07.5") # Middle of the day 7 March 2024
#> [1] 2024-03-07.5

Properties of a `datey`

The $ operator extracts components of a datey:

t <- mid_day(2024, 3, 7)
t$year
#> [1] 2024
t$month
#> [1] 3
t$day
#> [1] 7
t$day_fraction
#> [1] 0.5

If you need several components at once, it is more efficient to use to_ymdf() instead:

to_ymdf(t)
#> $year
#> [1] 2024
#> 
#> $month
#> [1] 3
#> 
#> $day
#> [1] 7
#> 
#> $day_fraction
#> [1] 0.5

as.double() converts to a fractional calendar year; as.integer() gives the calendar year:

as.double(t)
#> [1] 2024.182
as.integer(t)
#> [1] 2024

is_start_day() and is_mid_day() test the position within the day. Note that end_day() produces a datey at the start of the following day, so it tests as is_start_day():

is_start_day(start_day(2024, 3, 7)) # TRUE
#> [1] TRUE
is_mid_day(mid_day(2024, 3, 7))     # TRUE
#> [1] TRUE
is_start_day(end_day(2024, 3, 7))   # TRUE because end = start of next day
#> [1] TRUE
is_mid_day(datey(2024, 3, 7, 0.25)) # FALSE
#> [1] FALSE

Creating a `durationy`

durationys typically arise as datey differences:

dob <- start_day(as.Date("1965-09-12"))
dod <- mid_day(2024, 3, 7)
age <- dod - dob
age
#> [1] 58.485804 yr

You can create them explicitly using durationy(), which accepts a number of years:

durationy(1)      # One year
#> [1] 1 yr
durationy(0.5)    # Half a year
#> [1] 0.5 yr
durationy(-2.5)   # Two and a half years in the past
#> [1] −2.5 yr

And you can convert them back to numerics using as.double(), which gives the duration as years, and as.integer(), which truncates toward zero:

as.double(age)
#> [1] 58.4858
as.integer(age)   # Whole years only
#> [1] 58

Comparisons and arithmetic

A number of arithmetic operations are available for datey, durationy and datey_interval.

Beware that not all combinations are valid because, for instance, it doesn’t make sense to add two dates together.

The table below summarises the valid arithmetic and comparison operations. All arithmetic is carried out as exact integer arithmetic on the underlying click counts, so the results are exact and associative.

Left	Op	Right	Result
`datey`	`-`	`datey`	`durationy`
`datey`	`+ -`	`durationy`	`datey`
`durationy`	`+`	`datey`	`datey`
`durationy`	`+ -`	`durationy`	`durationy`
`datey`	`== != < <= > >=`	`datey`	`logical`
`durationy`	`== != < <= > >=`	`durationy`	`logical`
`datey`	`%to%`	`datey`	`datey_interval`
`datey_interval`	`== !=`	`datey_interval`	`logical`
`datey_interval`	`%includes%`	`datey`	`logical`
`datey_interval`	`&`	`datey_interval`	`datey_interval`

start  <- start_day(2000, 1, 1)
one_yr <- durationy(1)
qtr_yr <- durationy(0.25)

start + one_yr    # One year later
#> [1] 2001-01-01.0
start - qtr_yr    # Quarter of a year earlier
#> [1] 1999-10-01.75

one_yr - qtr_yr   # Three quarters of a year
#> [1] 0.75 yr
one_yr + qtr_yr
#> [1] 1.25 yr

datey(2024) < datey(2025)     # TRUE
#> [1] TRUE
durationy(1) > durationy(0.5) # TRUE
#> [1] TRUE

You can also do mixed arithmetic with datey and durationy and numbers, in which case dateys and durationys are first converted to doubles:

identical(datey(2000) + 25, 2025)     # TRUE
#> [1] TRUE
identical(durationy(2) * 0.05, 0.10)  # TRUE
#> [1] TRUE

`datey_interval` – representing a time period

A datey_interval is a half-open [start, end) interval. Create one with datey_interval() or the %to% operator:

a  <- start_day(2024, 1, 1)
b  <- start_day(2025, 1, 1)
interval <- a %to% b
interval
#> [1] [2024-01-01.0, 2025-01-01.0)

The $start, $end and $duration properties extract the interval’s components:

interval$start
#> [1] 2024-01-01.0
interval$end
#> [1] 2025-01-01.0
interval$duration
#> [1] 1 yr

durationy() accepts a datey_interval directly:

durationy(interval)
#> [1] 1 yr

Interval membership testing

%includes% tests whether a datey falls inside the interval. The interval includes its start and excludes its end:

interval %includes% a                     # TRUE  -- start is included
#> [1] TRUE
interval %includes% b                     # FALSE -- end is excluded
#> [1] FALSE
interval %includes% mid_day(2024, 6, 15)  # TRUE
#> [1] TRUE

Interval properties

is_proper() returns TRUE when start ≤ end; is_collapsed() returns TRUE when start ≥ end. A point interval [a, a) is both proper and collapsed (it contains no time):

is_proper(interval)    # TRUE because start <= end
#> [1] TRUE
is_collapsed(interval) # FALSE because start < end
#> [1] FALSE

point <- a %to% a      # Empty (point) interval
is_proper(point)       # TRUE because a <= a
#> [1] TRUE
is_collapsed(point)    # TRUE because a >= a
#> [1] TRUE

Intersection

The & operator returns the intersection of two datey_intervals. This is the most direct way to compute the overlap of two time periods:

period    <- start_day(2023, 7, 1) %to% end_day(2024, 6, 30)
year_2024 <- start_day(2024, 1, 1) %to% end_day(2024, 12, 31)

overlap <- period & year_2024
overlap
#> [1] [2024-01-01.0, 2024-07-01.0)

overlap$duration  # exposure in calendar year 2024, in years
#> [1] 0.497268 yr

NA values

Throughout the datey package, NA will cause an error when used where a datey_, durationy_ or datey_interval_ is expected. This is because the type of NA is logical. which has no meaningful date or duration interpretation therefore potentially indicates user error.

If you want an NA value with a datey system type, use the explicit forms NA_datey_, NA_durationy_ or NA_datey_interval_ as appropriate.

is.na() and anyNA() work as expected:

is.na(NA_datey_)
#> [1] TRUE
anyNA(c(datey(2000), NA_datey_, datey(2024)))
#> [1] TRUE

By default, out-of-range inputs stop execution. With strict = FALSE they become NA instead:

datey(999.9, strict = FALSE)           # Outside [1000,3000]: NA
#> [1] <NA>
start_day(2000, 0, 12, strict = FALSE) # Invalid month: NA
#> [1] <NA>
mid_day(2001, 2, 29, strict = FALSE)   # Invalid day (given year and month): NA
#> [1] <NA>
durationy(2000.1, strict = FALSE)      # exceeds 2000-year limit: NA
#> [1] <NA>

NA values propagate through arithmetic:

start_day(2024, 1, 1) + NA_durationy_
#> [1] <NA>

Sequences and statistics

seq(), min(), max(), range() and mean() all work on datey and durationy vectors:

dates <- c(datey(2021), datey(2022), datey(2023))

min(dates)
#> [1] 2021-01-01.0
max(dates)
#> [1] 2023-01-01.0
mean(dates)
#> [1] 2022-01-01.0

seq(from = datey(2020), to = datey(2024), by = durationy(2))
#> [1] 2020-01-01.0 2022-01-01.0 2024-01-01.0

Even though datey_interval stores the start and the end of a time interval, it too is atomic, which means that datey_intervals can be stored in a single vector without any additional special handling.↩︎
Even though the Date type is not designed for fractional dates, it typically uses floating point under the covers, and can unintentionally end up with a fractional value e.g. by taking a mean of Dates. For this reason, a day_fraction argument is always required for a Date.↩︎