thr3ads.net - R help - [R] Problem with timeSequence {timeDate}

If this information is useful, please help other people find it:
Share via:

Benjamin B.

2011-Jan-06 09:30 UTC

[R] Problem with timeSequence {timeDate} - wrong end date

Dear help-list,

I have a problem with timeSequence {timeDate}.

When I use it like
> timeSequence(from = "2008-01-01", to = "2010-12-13", by
= "1 month")GMT
 [1] [2008-01-01] [2008-02-01] [2008-03-01] [2008-04-01] [2008-05-01]
[2008-06-01] [2008-07-01] [2008-08-01] [2008-09-01] [2008-10-01]
[2008-11-01]
[12] [2008-12-01] [2009-01-01] [2009-02-01] [2009-03-01] [2009-04-01]
[2009-05-01] [2009-06-01] [2009-07-01] [2009-08-01] [2009-09-01]
[2009-10-01]
[23] [2009-11-01] [2009-12-01] [2010-01-01] [2010-02-01] [2010-03-01]
[2010-04-01] [2010-05-01] [2010-06-01] [2010-07-01] [2010-08-01]
[2010-09-01]
[34] [2010-10-01] [2010-11-01] [2010-12-01]

The result is as expected: a list of dates with all dates smaller then the
"to" date.

But somehow it behaves strange when I use it with a different starting date:
> test <- timeSequence(from = "2008-01-15", to =
"2010-12-13", by = "1month")
GMT
 [1] [2008-01-15] [2008-02-15] [2008-03-15] [2008-04-15] [2008-05-15]
[2008-06-15] [2008-07-15] [2008-08-15] [2008-09-15] [2008-10-15]
[2008-11-15]
[12] [2008-12-15] [2009-01-15] [2009-02-15] [2009-03-15] [2009-04-15]
[2009-05-15] [2009-06-15] [2009-07-15] [2009-08-15] [2009-09-15]
[2009-10-15]
[23] [2009-11-15] [2009-12-15] [2010-01-15] [2010-02-15] [2010-03-15]
[2010-04-15] [2010-05-15] [2010-06-15] [2010-07-15] [2010-08-15]
[2010-09-15]
[34] [2010-10-15] [2010-11-15] [2010-12-15]

In this case the last calculated date is obviously LARGER than the
"to"
date:
> as.Date(test[length(test)]) < "2010-12-13"[1] FALSE

This seem to occur also with other parameters:
> timeSequence(from = "1999-12-12", to = "2000-06-08", by
= "2 months")GMT
[1] [1999-12-12] [2000-02-12] [2000-04-12] [2000-06-12]
> timeSequence(from = as.Date("1999-12-12"), to =
as.Date("2000-06-08"), by= "2 months")
GMT
[1] [1999-12-12] [2000-02-12] [2000-04-12] [2000-06-12]

Am I missing something essential in using timeSequence?
Is this behavior wanted (then I don't get why it should...)?
Is there a better way to get those dates?

Thanks for reading and greetings,

Benjamin


Benjamin B.
Hamburg, Germany

Uwe Ligges

2011-Jan-06 12:57 UTC

head link

[R] Problem with timeSequence {timeDate} - wrong end date

Please report bugs in contributed packages to the corresponding package 
maintainer.

Uwe  Ligges


On 06.01.2011 10:30, Benjamin B. wrote:> Dear help-list,
>
> I have a problem with timeSequence {timeDate}.
>
> When I use it like
>
>> timeSequence(from = "2008-01-01", to =
"2010-12-13", by = "1 month")
> GMT
>   [1] [2008-01-01] [2008-02-01] [2008-03-01] [2008-04-01] [2008-05-01]
> [2008-06-01] [2008-07-01] [2008-08-01] [2008-09-01] [2008-10-01]
> [2008-11-01]
> [12] [2008-12-01] [2009-01-01] [2009-02-01] [2009-03-01] [2009-04-01]
> [2009-05-01] [2009-06-01] [2009-07-01] [2009-08-01] [2009-09-01]
> [2009-10-01]
> [23] [2009-11-01] [2009-12-01] [2010-01-01] [2010-02-01] [2010-03-01]
> [2010-04-01] [2010-05-01] [2010-06-01] [2010-07-01] [2010-08-01]
> [2010-09-01]
> [34] [2010-10-01] [2010-11-01] [2010-12-01]
>
> The result is as expected: a list of dates with all dates smaller then the
> "to" date.
>
> But somehow it behaves strange when I use it with a different starting
date:
>
>> test<- timeSequence(from = "2008-01-15", to =
"2010-12-13", by = "1
> month")
> GMT
>   [1] [2008-01-15] [2008-02-15] [2008-03-15] [2008-04-15] [2008-05-15]
> [2008-06-15] [2008-07-15] [2008-08-15] [2008-09-15] [2008-10-15]
> [2008-11-15]
> [12] [2008-12-15] [2009-01-15] [2009-02-15] [2009-03-15] [2009-04-15]
> [2009-05-15] [2009-06-15] [2009-07-15] [2009-08-15] [2009-09-15]
> [2009-10-15]
> [23] [2009-11-15] [2009-12-15] [2010-01-15] [2010-02-15] [2010-03-15]
> [2010-04-15] [2010-05-15] [2010-06-15] [2010-07-15] [2010-08-15]
> [2010-09-15]
> [34] [2010-10-15] [2010-11-15] [2010-12-15]
>
> In this case the last calculated date is obviously LARGER than the
"to"
> date:
>
>> as.Date(test[length(test)])<  "2010-12-13"
> [1] FALSE
>
> This seem to occur also with other parameters:
>
>> timeSequence(from = "1999-12-12", to =
"2000-06-08", by = "2 months")
> GMT
> [1] [1999-12-12] [2000-02-12] [2000-04-12] [2000-06-12]
>
>> timeSequence(from = as.Date("1999-12-12"), to =
as.Date("2000-06-08"), by
> = "2 months")
> GMT
> [1] [1999-12-12] [2000-02-12] [2000-04-12] [2000-06-12]
>
> Am I missing something essential in using timeSequence?
> Is this behavior wanted (then I don't get why it should...)?
> Is there a better way to get those dates?
>
> Thanks for reading and greetings,
>
> Benjamin
>
>
> Benjamin B.
> Hamburg, Germany
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

Joshua Wiley

2011-Jan-06 13:27 UTC

head link

[R] Problem with timeSequence {timeDate} - wrong end date

Dear Benjamin,

timeSequence() ultimately is relying on seq.POSIXt().  If you look at
the last paragraph of the "Details" section in ?seq.POSIXt it seems to
basically indicate that using by = "month" just sequences through the
months and the day is only changed if it is invalid for a particular
month (the relevant lines are around 88--100 in the seq.POSIXt code).

It is fairly straightforward ensure the output is less than "to".
Here is one option:

test <- timeSequence(from = "2008-01-01", to =
"2010-12-13", by = "1 month")
test <- test[as.Date(test) < as.Date("2010-12-13")]
test

Cheers,

Josh


On Thu, Jan 6, 2011 at 1:30 AM, Benjamin B. <benj.bad.ac at
googlemail.com> wrote:> Dear help-list,
>
> I have a problem with timeSequence {timeDate}.
>
> When I use it like
>
>> timeSequence(from = "2008-01-01", to =
"2010-12-13", by = "1 month")
> GMT
> ?[1] [2008-01-01] [2008-02-01] [2008-03-01] [2008-04-01] [2008-05-01]
> [2008-06-01] [2008-07-01] [2008-08-01] [2008-09-01] [2008-10-01]
> [2008-11-01]
> [12] [2008-12-01] [2009-01-01] [2009-02-01] [2009-03-01] [2009-04-01]
> [2009-05-01] [2009-06-01] [2009-07-01] [2009-08-01] [2009-09-01]
> [2009-10-01]
> [23] [2009-11-01] [2009-12-01] [2010-01-01] [2010-02-01] [2010-03-01]
> [2010-04-01] [2010-05-01] [2010-06-01] [2010-07-01] [2010-08-01]
> [2010-09-01]
> [34] [2010-10-01] [2010-11-01] [2010-12-01]
>
> The result is as expected: a list of dates with all dates smaller then the
> "to" date.
>
> But somehow it behaves strange when I use it with a different starting
date:
>
>> test <- timeSequence(from = "2008-01-15", to =
"2010-12-13", by = "1
> month")
> GMT
> ?[1] [2008-01-15] [2008-02-15] [2008-03-15] [2008-04-15] [2008-05-15]
> [2008-06-15] [2008-07-15] [2008-08-15] [2008-09-15] [2008-10-15]
> [2008-11-15]
> [12] [2008-12-15] [2009-01-15] [2009-02-15] [2009-03-15] [2009-04-15]
> [2009-05-15] [2009-06-15] [2009-07-15] [2009-08-15] [2009-09-15]
> [2009-10-15]
> [23] [2009-11-15] [2009-12-15] [2010-01-15] [2010-02-15] [2010-03-15]
> [2010-04-15] [2010-05-15] [2010-06-15] [2010-07-15] [2010-08-15]
> [2010-09-15]
> [34] [2010-10-15] [2010-11-15] [2010-12-15]
>
> In this case the last calculated date is obviously LARGER than the
"to"
> date:
>
>> as.Date(test[length(test)]) < "2010-12-13"
> [1] FALSE
>
> This seem to occur also with other parameters:
>
>> timeSequence(from = "1999-12-12", to =
"2000-06-08", by = "2 months")
> GMT
> [1] [1999-12-12] [2000-02-12] [2000-04-12] [2000-06-12]
>
>> timeSequence(from = as.Date("1999-12-12"), to =
as.Date("2000-06-08"), by
> = "2 months")
> GMT
> [1] [1999-12-12] [2000-02-12] [2000-04-12] [2000-06-12]
>
> Am I missing something essential in using timeSequence?
> Is this behavior wanted (then I don't get why it should...)?
> Is there a better way to get those dates?
>
> Thanks for reading and greetings,
>
> Benjamin
>
>
> Benjamin B.
> Hamburg, Germany
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


-- 
Joshua Wiley
Ph.D. Student, Health Psychology
University of California, Los Angeles
http://www.joshuawiley.com/

Joshua Wiley

2011-Jan-06 14:47 UTC

head link

[R] Problem with timeSequence {timeDate} - wrong end date

On Thu, Jan 6, 2011 at 5:27 AM, Joshua Wiley <jwiley.psych at gmail.com>
wrote:> timeSequence() ultimately is relying on seq.POSIXt().  If you look at
My apologies, I spoke nonsense---timeSequence() does NOT rely on
seq.POSIXt().  The timeDate package has its own method defined for
seq() which is what gets dispatched.  Still, the behavior is similar:
> timeSequence(from = "2010-01-15", to = "2010-02-01", by
= "1 month")GMT
[1] [2010-01-15] [2010-02-15]>
> seq.POSIXt(from = as.POSIXct("2010-01-15"),+   to = as.POSIXct("2010-02-01"), by = "1 month")
[1] "2010-01-15 PST" "2010-02-15 PST"


which is not surprising because the code is responsible for this
behavior is similar:

###### seq.timeDate* ######
else if (valid == 6) {
  if (missing(to)) {
    mon <- seq.int(r1$mon, by = by, length.out = length.out)
  }
  else {
    to <- as.POSIXlt(to)
    mon <- seq.int(r1$mon, 12 * (to$year - r1$year) +
      to$mon, by)
  }
  r1$mon <- mon
  r1$isdst <- -1
  res <- as.POSIXct(r1)
}

###### seq.POSIXt ######
else if (valid == 6L) {
  if (missing(to)) {
    mon <- seq.int(r1$mon, by = by, length.out = length.out)
  }
  else {
    to <- as.POSIXlt(to)
    mon <- seq.int(r1$mon, 12 * (to$year - r1$year) +
      to$mon, by)
  }
  r1$mon <- mon
  r1$isdst <- -1
  res <- as.POSIXct(r1)
}

and "res" is the object returned in both cases (I believe).

My system:
R version 2.12.1 (2010-12-16)
Platform: i486-pc-linux-gnu (32-bit)

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
 [5] LC_MONETARY=C              LC_MESSAGES=en_US.UTF-8
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
 [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] timeDate_2130.91

Martin Maechler

2011-Jan-07 14:59 UTC

head link

[R] Problem with timeSequence {timeDate} - wrong end date

>>>>> Joshua Wiley <jwiley.psych at gmail.com>
>>>>>     on Thu, 6 Jan 2011 06:47:43 -0800 writes:
    > On Thu, Jan 6, 2011 at 5:27 AM, Joshua Wiley <jwiley.psych at
gmail.com> wrote:
    >> timeSequence() ultimately is relying on seq.POSIXt().  If you look
at

    > My apologies, I spoke nonsense---timeSequence() does NOT rely on
    > seq.POSIXt().  The timeDate package has its own method defined for
    > seq() which is what gets dispatched.  Still, the behavior is similar:

    >> timeSequence(from = "2010-01-15", to =
"2010-02-01", by = "1 month")
    > GMT
    > [1] [2010-01-15] [2010-02-15]
    >> 
    >> seq.POSIXt(from = as.POSIXct("2010-01-15"),
    > +   to = as.POSIXct("2010-02-01"), by = "1 month")
    > [1] "2010-01-15 PST" "2010-02-15 PST"


    > which is not surprising because the code is responsible for this
    > behavior is similar:

    > ###### seq.timeDate* ######
    > else if (valid == 6) {
    > if (missing(to)) {
    > mon <- seq.int(r1$mon, by = by, length.out = length.out)
    > }
    > else {
    > to <- as.POSIXlt(to)
    > mon <- seq.int(r1$mon, 12 * (to$year - r1$year) +
    > to$mon, by)
    > }
    > r1$mon <- mon
    > r1$isdst <- -1
    > res <- as.POSIXct(r1)
    > }

    > ###### seq.POSIXt ######
    > else if (valid == 6L) {
    > if (missing(to)) {
    > mon <- seq.int(r1$mon, by = by, length.out = length.out)
    > }
    > else {
    > to <- as.POSIXlt(to)
    > mon <- seq.int(r1$mon, 12 * (to$year - r1$year) +
    > to$mon, by)
    > }
    > r1$mon <- mon
    > r1$isdst <- -1
    > res <- as.POSIXct(r1)
    > }

    > and "res" is the object returned in both cases (I believe).

Thank you, Joshua.

Note that "R 2.12.1 patched" and "R 2.13.0 unstable"
have this fixed (since yesterday) and hence will no longer
"overshoot".

Also, the next release of the  timeDate  package will have it
fixed.


Martin Maechler, ETH Zurich



    > My system:
    > R version 2.12.1 (2010-12-16)
    > Platform: i486-pc-linux-gnu (32-bit)

    > locale:
    > [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
    > [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
    > [5] LC_MONETARY=C              LC_MESSAGES=en_US.UTF-8
    > [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
    > [9] LC_ADDRESS=C               LC_TELEPHONE=C
    > [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

    > attached base packages:
    > [1] stats     graphics  grDevices utils     datasets  methods   base

    > other attached packages:
    > [1] timeDate_2130.91

    > ______________________________________________
    > R-help at r-project.org mailing list
    > https://stat.ethz.ch/mailman/listinfo/r-help
    > PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
    > and provide commented, minimal, self-contained, reproducible code.

Joshua Wiley

2011-Jan-07 15:54 UTC

head link

[R] Problem with timeSequence {timeDate} - wrong end date

Thank you, Martin (and all of R core) for all your work making such
wonderful software available.  Not only is it practical, using R,
going through the source code and the documentation has been a huge
learning experience for me.  I am very grateful to you and everyone
else for being so generous with your time.

Josh

On Fri, Jan 7, 2011 at 6:59 AM, Martin Maechler
<maechler at stat.math.ethz.ch> wrote:> Thank you, Joshua.
>
> Note that "R 2.12.1 patched" and "R 2.13.0 unstable"
> have this fixed (since yesterday) and hence will no longer
"overshoot".
>
> Also, the next release of the ?timeDate ?package will have it
> fixed.
>
>
> Martin Maechler, ETH Zurich

Maybe Matching Threads

Search for more apparently analagous threads

R help - Jan 2011 - Problem with timeSequence {timeDate} - wrong end date

[R] Problem with timeSequence {timeDate} - wrong end date

[R] Problem with timeSequence {timeDate} - wrong end date

[R] Problem with timeSequence {timeDate} - wrong end date

[R] Problem with timeSequence {timeDate} - wrong end date

[R] Problem with timeSequence {timeDate} - wrong end date

[R] Problem with timeSequence {timeDate} - wrong end date

Maybe Matching Threads