[R-SIG-Finance] Non unique timestamp in zoo object

[Jeff Ryan]

Jatin,

quantmod charts need non-unique values at present.  I suspect that
will always be the case, as the x-index relies on time.

xts can handle merging with non-unique timestamps, but keep in mind
that many operations would be quite ill-defined.  Merging will work,
but Ops methods (+-/* etc) will make little sense if you aren't
cognisant of what is actually being performed. diff() and lag() will
also be problematic.

Aside from that, the myriad of documentation available on zoo (great),
xts (less great, but good), and using the above and R in finance
(check the list archives... learn the list archives) is quite
impressive.

Obviously the caliber of the reply on this list is awe inspiring as
well, but asking questions should really be a last resort, and only if
they haven't been asked before.

Jeff


On Tue, Sep 1, 2009 at 12:22 PM, Gabor
Grothendieck<ggrothendieck at gmail.com> wrote:
> Note that for any calculation that involves merge.zoo (which
> is most calculations since even z1 + z2 can do a
> merge underneath) one does require unique times in zoo.
> read.zoo can read in times that are not unique but ultimately
> you will want to make them unique for most operations.
>
> The aggregate= argument of read.zoo provides a way to
> collapse duplicates and there are additional facilities in the
> devel version of zoo.
>
> As indicated in my prior post in this thread zoo FAQ #1
> has further discussion.
>
> On Tue, Sep 1, 2009 at 1:13 PM, Brian G. Peterson<brian at braverock.com> wrote:
>> jatin patni wrote:
>>>
>>> I'm trying to create a zoo object from a tick data file(sample below)
>>> which
>>> contains timestamps only.
>>> Filename: "20080202.trd"
>>> 1|BHARTIARTL|EQ|18:15:05|600|1
>>> 2|BHARTIARTL|EQ|18:15:05|600|99
>>> 3|GLENMARK|EQ|18:15:05|238.1|5
>>> 4|HINDALCO|EQ|18:15:05|43.75|100
>>> 5|BHARTIARTL|EQ|18:15:05|600|1
>>> 6|BHEL|EQ|18:15:05|1100|11
>>> 7|HINDALCO|EQ|18:15:06|43.2|1
>>> 8|CHAMBLFERT|EQ|18:15:06|46|10
>>> 9|CHAMBLFERT|EQ|18:15:06|46|90
>>> 10|BAJAUTOFIN|EQ|18:15:06|80|100
>>>
>>> As you can see that timestamps are not unique even at microsecond levels.
>>> Inorder to create a zoo/xts object to be used with quantmod later, How do
>>> I:-
>>> 1)Do I have to merge two rows with the same timestamp, symbols, price but
>>> different volumes.(Do I need to do this or is there some other way
>>> possible)
>>> 2)Append date to the timestamp while creating the zoo/xts object
>>>
>>> I'm trying to explore the 'format' argument, and POSIX and chron but
>>> without
>>> much success.
>>>
>>
>> Your target should be POSIXct, as that is what xts uses internally, given
>> that your stated goal is to use xts and quantmod.
>>
>> ?as.POSIXct
>> ?format.POSIXct
>>
>> These look like one second bars, not microsecond bars.  To see microseconds,
>> if you have them:
>>
>> options(digits.secs=6)
>>
>> zoo doesn't require unique timestamps, just ordered ones.  In this case,
>> your trades are ordered in the files.  You might need to append an order in
>> the microseconds if you really only have second bars.  It's also not
>> necessary to have unique timestamps, but you'll be happier if you do.
>>
>> Regards,
>>
>>  - Brian
>>
>>
>> --
>> Brian G. Peterson
>> http://braverock.com/brian/
>> Ph: 773-459-4973
>> IM: bgpbraverock
>>
>> _______________________________________________
>> 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.
>>
>
> _______________________________________________
> 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