Truncating dates (and other date-time manipulations)

On Thu, Sep 11, 2008 at 11:00 AM, Gabor Grothendieck
<ggrothendieck at gmail.com> wrote:

See ?cut.Date

In the zoo package see:
?as.yearmon
?as.yearqtr
?aggregate.zoo

and the many examples in:
?plot.zoo
?xyplot.zoo
as well as the three zoo vignettes.

Also in the xts package look at
?to.period

For regularly spaced series the tis package supports a wide
variety of time bases and can convert among them.

There are as.zoo.tis and as.tis.zoo routines in zoo and tis; also,
xts is a subclass of zoo so all these packages can work together.

Ok, lets take a concrete example:

dates <- structure(c(8516, 8544, 8568, 8596, 8609, 8666, 8701, 8750, 8754,
8798, 8811, 8817, 8860, 8873, 8918, 8931, 8966, 9020, 9034, 9056
), class = "Date")

range(dates)

[1] "1993-04-26" "1994-10-18"

I want to be able to say: give me a monthly time series that spans
this range - i.e. it should start on 1993-04-01 and continue to
1994-11-01.  I should equally easily be able to say give me a yearly
time series, or every 2 months or every 3 months, or 2 weeks etc.

I don't think that cut.Date helps because I want to make a new series,
not divide up an existing one, similarly with to.period.  as.yearmon,
as.yearqtr etc, might be helpful but I'd need to stitch them together
myself and they don't return dates so I'd have to convert back for
plotting.   plot.zoo doesn't help because all the examples are regular
time series.

Hadley,

What's wrong with:

dates <- structure(c(8516, 8544, 8568, 8596, 8609, 8666, 8701, 8750,
                     8754, 8798, 8811, 8817, 8860, 8873, 8918, 8931,
                     8966, 9020, 9034, 9056), class = "Date")

as.Date(cut.Date(dates, "year"))

[1] "1993-01-01" "1993-01-01" "1993-01-01" "1993-01-01" "1993-01-01"
 [6] "1993-01-01" "1993-01-01" "1993-01-01" "1993-01-01" "1994-01-01"
[11] "1994-01-01" "1994-01-01" "1994-01-01" "1994-01-01" "1994-01-01"
[16] "1994-01-01" "1994-01-01" "1994-01-01" "1994-01-01" "1994-01-01"

as.Date(cut.Date(dates, "month"))

[1] "1993-04-01" "1993-05-01" "1993-06-01" "1993-07-01" "1993-07-01"
 [6] "1993-09-01" "1993-10-01" "1993-12-01" "1993-12-01" "1994-02-01"
[11] "1994-02-01" "1994-02-01" "1994-04-01" "1994-04-01" "1994-06-01"
[16] "1994-06-01" "1994-07-01" "1994-09-01" "1994-09-01" "1994-10-01"

as.Date(cut.Date(dates, "2 months"))

[1] "1993-04-01" "1993-04-01" "1993-06-01" "1993-06-01" "1993-06-01"
 [6] "1993-08-01" "1993-10-01" "1993-12-01" "1993-12-01" "1994-02-01"
[11] "1994-02-01" "1994-02-01" "1994-04-01" "1994-04-01" "1994-06-01"
[16] "1994-06-01" "1994-06-01" "1994-08-01" "1994-08-01" "1994-10-01"

as.Date(cut.Date(dates, "3 months"))

[1] "1993-04-01" "1993-04-01" "1993-04-01" "1993-07-01" "1993-07-01"
 [6] "1993-07-01" "1993-10-01" "1993-10-01" "1993-10-01" "1994-01-01"
[11] "1994-01-01" "1994-01-01" "1994-04-01" "1994-04-01" "1994-04-01"
[16] "1994-04-01" "1994-07-01" "1994-07-01" "1994-07-01" "1994-10-01"

as.Date(cut.Date(dates, "quarter"))

[1] "1993-04-01" "1993-04-01" "1993-04-01" "1993-07-01" "1993-07-01"
 [6] "1993-07-01" "1993-10-01" "1993-10-01" "1993-10-01" "1994-01-01"
[11] "1994-01-01" "1994-01-01" "1994-04-01" "1994-04-01" "1994-04-01"
[16] "1994-04-01" "1994-07-01" "1994-07-01" "1994-07-01" "1994-10-01"

as.Date(cut.Date(dates, "2 weeks"))

[1] "1993-04-26" "1993-05-24" "1993-06-07" "1993-07-05" "1993-07-19"
 [6] "1993-09-13" "1993-10-25" "1993-12-06" "1993-12-20" "1994-01-31"
[11] "1994-02-14" "1994-02-14" "1994-03-28" "1994-04-11" "1994-05-23"
[16] "1994-06-06" "1994-07-18" "1994-09-12" "1994-09-26" "1994-10-10"


You can of course, format the dates as needed for aesthetic purposes.

Isn't that what you seem to be looking for?

BTW, seq.Date() supports the same time intervals designations.

HTH,

Marc Schwartz