[R-SIG-Finance] Concatenating time series objects?

[Jeff Ryan]

Hi Marc,

rbind and cbind are particularly difficult in R.

They are not true generic functions, but rather calls to .Internal
code.  This mean that the dispatch is very ... different :)

You can try calling rbind.xts directly -- that may indeed solve your problem.

This issue has been addressed in the new xts, which is being prepared
for release as I type this.  Basically I have added custom xts code (C
code) to bypass the dispatching quirks that arise from the entire
rbind->rbind.xts->rbind.zoo calling structure.

Again, the look and feel of zoo will remain intact, but the underlying
code is going to be different.  Precisely to address issues like this
that have been arising.

Jeff


On Tue, Dec 2, 2008 at 12:48 AM, Marc Delvaux <mdelvaux at gmail.com> wrote:
> I should have been more specific, rbind and cbind are the first
> solution I tried, but then the resulting object doesn't seem to be an
> xts :-(  It mostly look like one, but str reports it as such and the
> time information seems lost, at least to my eye.  I also tried merge,
> but this results in essentially a multivariate object, not in
> concatenation intime.
>
> see below for a toy example
>
>> seq1 <- xts(1:10,order.by=as.Date(a))
>> seq2 <- xts(1:10,order.by=as.Date(b))  << a and b are two consecutive sequences
>> str(seq1)
> An 'xts' object from 2008-12-02 to 2008-12-12 containing:
>  Data: int [1:11] 1 2 3 4 5 6 7 8 9 10 ...
>  Indexed by: Class 'Date'  int [1:11] 14215 14216 14217 14218 14219
> 14220 14221 14222 14223 14224 ...
>  xts Attributes:
>  NULL
>> seq <- rbind(seq1, seq2)
> Warning message:
> number of columns of result
>        is not a multiple of vector length (arg 2) in: rbind(1, seq1, seq2)
>> str(seq)
>  int [1:2, 1:11] 1 1 2 2 3 3 4 4 5 5 ...
>  - attr(*, "dimnames")=List of 2
>  ..$ : chr [1:2] "seq1" "seq2"
>  ..$ : NULL
>> seq <- cbind(seq1, seq2)
> Warning message:
> number of rows of result
>        is not a multiple of vector length (arg 2) in: cbind(1, seq1, seq2)
>> str(seq)
>  int [1:11, 1:2] 1 2 3 4 5 6 7 8 9 10 ...
>  - attr(*, "dimnames")=List of 2
>  ..$ : NULL
>  ..$ : chr [1:2] "seq1" "seq2"
>>
>
>
> On Mon, Dec 1, 2008 at 10:01 PM, Jeff Ryan <jeff.a.ryan at gmail.com> wrote:
>> ?rbind
>>
>> Jeff
>>
>> On Mon, Dec 1, 2008 at 11:55 PM, Marc Delvaux <mdelvaux at gmail.com> wrote:
>>> I am trying to do an incremental download of time series financial
>>> data, with say 5 new points every time I do an download.  I am trying
>>> to find a simple way to perform concatenation but I cannot seem to
>>> find an easy way to do that.  I did find a way using
>>> xts(c(coredata(seq1),coredata(seq2)),
>>> order.by=c(time(seq1),time(seq2))), but this loses some information
>>> that was present in seq1, and I was thinking there should be a simpler
>>> way.
>>>
>>> _______________________________________________
>>> R-SIG-Finance at stat.math.ethz.ch mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>>> -- Subscriber-posting only.
>>> -- If you want to post, subscribe first.
>>>
>>
>>
>>
>> --
>> Jeffrey Ryan
>> jeffrey.ryan at insightalgo.com
>>
>> ia: insight algorithmics
>> www.insightalgo.com
>>
>



-- 
Jeffrey Ryan
jeffrey.ryan at insightalgo.com

ia: insight algorithmics
www.insightalgo.com