[R-SIG-Finance] Luxor strategy (quantstrat) - Why are successive short (and long) trades happening ?

[Joshua Ulrich]

Hi Pablo,

I just looked into this with Brian Peterson and Guy Yollin.  It
appears a bug was accidentally introduced in r1594.  This should be
fixed in r1625.  Thanks for the report, and sorry the fix took so
long.

Best,
--
Joshua Ulrich  |  about.me/joshuaulrich
FOSS Trading  |  www.fosstrading.com


On Sat, May 10, 2014 at 5:02 PM, Pablo Rios <pablo.javier.rios at gmail.com> wrote:
> I am also unclear about results Amarjit.
>
> Furthermore, why is the position = 0.0 after entering long at 2002-10-30
> 13:00:00 (see flat Positionfill and CumPL lines) ? There's a long signal at
> 2002-10-30 12:00:00 (see mktdata xts object) and the condition to enter long
> is satisfied (i.e.: .threshold) at 13:00:00 (see order book). Is this
> another anomaly with blotter/quantstrat ?
>
> Pablo
>
>
>
> On Sat, May 10, 2014 at 3:17 PM, amarjit chandhial
> <a.chandhial at btinternet.com> wrote:
>>
>>
>>
>> I am unclear. There are 3 different results, which of these is correct ?
>>
>>
>> (a) Humme, Peterson pp.34/78
>> http://www.rinfinance.com/agenda/2013/workshop/Humme+Peterson.pdf
>>
>>
>>
>> (b) Guy Yollin pp.24/78
>>
>> https://4310b1a9-a-a8fb2076-s-sites.googlegroups.com/a/r-programming.org/home/files/quantstrat-IV.pdf?attachauth=ANoY7cojS8K93NCMEI2gr1eWPE6nFGkEZAncLma__qYUXHgRzbbQCi1zrOZa5DnAANVw6nGZU9ppV1s69und3U7_uErEyb18pGOyH0UJsXtCsndrvcZWR4fB4mIJePELsQcuf3ksoDg3w7JV0TH_kpR3NDGBiDYzX9f43piDIk6vhfK5JMK68K1K7yqCZcHZ0krZzhRJ9Wq3KIZt5-399IqQ-Eeytar1o2n-VpqgoBIefXlC5iT6rlM%3D&attredirects=0
>>
>>
>>
>> (c) The latest run of the demo, blotter(rev 1607), quantstrat(rev 1610) -
>> chart.Posn attached.
>>
>>
>>
>> Amarjit
>>
>>
>>
>> From: Joshua Ulrich <josh.m.ulrich at gmail.com>
>> To: Pablo Rios <pablo.javier.rios at gmail.com>
>> Cc: r-sig-finance <r-sig-finance at r-project.org>
>> Sent: Friday, 9 May 2014, 12:26
>>
>> Subject: Re: [R-SIG-Finance] Luxor strategy (quantstrat) - Why are
>> successive short (and long) trades happening ?
>>
>> On Thu, May 8, 2014 at 11:09 PM, Pablo Rios <pablo.javier.rios at gmail.com>
>> wrote:
>> > Thanks for your quick response Joshua.
>> >
>> > Does the code changes in r1609 intend to fix the successive short (or
>> > long)
>> > trades that I'm describing in my email, or only the Warning: stack
>> > imbalance
>> > in 'lapply' message ? Although this warning message is no longer
>> > reported,
>> > and the results of the Luxor strategy changed after r1609 (ex.:
>> > Net.Trading.PL value changed, among other variables), I'm still
>> > observing
>> > the same behaviour of successive short (or long) trades running the
>> > Luxor
>> > demo code with the GBP/USD demo data available in quantstrat.
>> >
>> r1609 only intended to fix the stack imbalance warning (as it says in
>> the commit log).
>>
>> > Moreover, if I run luxor.1.strategy.basic.R demo code with a longer
>> > GBP/USD
>> > time series, using 30 minutes bars as in the demo data (ex.: three
>> > years), I
>> > observed up to 5 successive short trades (i.e., Pos.Qty value of
>> > GBPUSD$posPL xts object equal to -500,000) and 4 successive long trades.
>> >
>> > Further, looking at the firstCross.c in r1609 I'm seeing in the
>> > switch(int_rel) statement that all comparisons are done by greater than
>> > ('>'). I don't know whether this is correct or not.
>> >
>> That was a careless error on my part.  r1610 corrects the comparisons.
>> Thank you very much for catching this.
>>
>> > Thanks for your support, I'm eager to finally adopt quantstrat !
>> >
>> > Pablo
>> >
>> >
>> > On Thu, May 8, 2014 at 8:09 AM, Joshua Ulrich <josh.m.ulrich at gmail.com>
>> > wrote:
>> >>
>> >> On Wed, May 7, 2014 at 7:05 PM, fc_11 <jyorio at gmail.com> wrote:
>> >> > i also am getting the "Warning: stack imbalance in 'lapply'," warning
>> >> > since
>> >> > the 1608 upgrade
>> >> >
>> >> Fixed in r1609.
>> >> --
>> >> Joshua Ulrich  |  about.me/joshuaulrich
>> >> FOSS Trading  |  www.fosstrading.com
>> >>
>> >> _______________________________________________
>> >> 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.
>>
>>
>