[R-SIG-Finance] quantstrat - macross demo problem

[Joshua Ulrich]

I just installed the most current versions of xts, blotter, and
quantstrat from R-forge.  I then ran the maCross.R demo after
uncommenting the two lines for short entries / exits.  I don't receive
any "cross through zero" warnings.  I can't replicate your issue.
--
Joshua Ulrich  |  FOSS Trading: www.fosstrading.com


On Mon, Feb 7, 2011 at 3:23 PM, Stephen Choularton
<stephen at organicfoodmarkets.com.au> wrote:
>
> Sorry, I must still not be making myself clear for which I apologiese.  I have downloaded current versions of xts and quantstrat and it all appears to be working.
>
> But it appears it is working just the same as before, that is with the ordering problem still in place.  I shouldn't have mentioned xts.  I guess the real question is has quantstrat been updated to sort out the problem:
>
> If you put the rules in the order of long enter then long exit, short enter and short exit the way the rules are processed the long open gets closed by the short open and the short close never gets found because of the crossing zero rule.
>
> Alexander Rudnev got to the bottom of the problem and Joshua Ulrich made the amendments and said they were committed but I still found the problem.
>
>
> Stephen Choularton Ph.D., FIoD
>
> 9999 2226
> 0413 545 182
>
>
> for insurance go to www.netinsure.com.au
> for markets go to www.organicfoodmarkets.com.au
>
> On 08/02/2011 7:57 AM, Jeffrey Ryan wrote:
>
> The issue is likely in the timing of your downloads.
>
> You can't install software that hasn't had its dependencies met.
> Likely your 'install' of quantstrat wasn't an install, failing when
> xts wasn't the right version.
>
> Getting the xts version required by quantstrat was what likely let the
> install (or load?) proceed. (btw, the xts bug was in the new column
> subsetting that was in .14 -- a quick patch - and didn't concern any
> other versions of xts)
>
> All works now?
>
> HTH
> Jeff
>
> On Mon, Feb 7, 2011 at 2:38 PM, Stephen Choularton
> <stephen at organicfoodmarkets.com.au> wrote:
>
> Sorry must not have made my question clear.  I downloaded the new quantstrat
> and when I tried to run it I got the error message.  Then I downloaded the
> new xts and it all worked again, but I thought that quantstrat had been
> fixed for the problem I ran into and have written about in this string.  If
> you put the rules in the order of long enter then long exit, short enter and
> short exit the way the rules are processed the long open gets closed by the
> short open and the short close never gets found because of the crossing zero
> rule.
>
> Alexander Rudnev got to the bottom of the problem and Joshua Ulrich made the
> amendments and said they were committed.   However, when I downloaded the
> current quantstrat I still found the problem.  I wondered if the current
> quantstrat does solve the problem, or if somehow the amendments have not got
> in their?
>
> Thanks.
>
> Stephen Choularton Ph.D., FIoD
> On 08/02/2011 5:40 AM, Brian G. Peterson wrote:
>
> On 02/07/2011 02:15 AM, Stephen Choularton wrote:
>
> Failed with error: ‘package 'xts' 0.7-6.11 was found, but >= 0.7.6.15 is
> required by 'quantstrat'’
>
> This is most likely your error.  Update xts first, then quantstrat. Other
> changes in xts fix a subsetting bug, and I think they were significant
> enough that I bumped the requirement for quantstrat as well.
>
> Alternately, use Rtools to build the binary package, and you can edit the
> DESCRIPTION file as needed to relax the constraints.
>
>  - Brian
>
> _______________________________________________
> R-SIG-Finance at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions
> should go.
>
>
> _______________________________________________
> R-SIG-Finance at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.