Ah - wait - I have an idea...

While I have no idea how negative numbers are stored in binary form in AFL (and 
C for that matter - only been coding assembly almost 2 decades ago), my guess 
is that when you view them as a positive number, you get highest possible 32 
bit number.

Thus by setting price = -1, price then is an extremely high number and the 
trade is rejected as funds are always insufficient (except for the case that my 
name is Scrooge McDuck...).

Correct?

--- In [email protected], "rise_t575" <ris...@...> wrote:
>
> 
> 
> 
> 
> 
> Tomasz,
> 
> While I have no idea what exactly does happen when I set sig.Price to -1 and 
> why this works, I can happily report that it *does* work - so thanks for the 
> help!
> 
> --- In [email protected], Tomasz Janeczko <groups@> wrote:
> >
> >   Hello,
> > 
> > It is simple. By default AmiBroker opens positions in RANKED order (more 
> > preferred trades first).
> > If at some point of going through the ranking it finds the trade that it 
> > can not open (for example
> > due to insufficient funds) it will NOT open LOWER ranked trades because it 
> > is undesirable to have
> > better candidates replaced by worse candidates when you run out of funds.
> > When requested position size is non-zero and position size shrinking is ON 
> > AmiBroker will attempt
> > to adjust pos size to lower value within user-defined constraints 
> > ("MinShares", Round lot size).
> > It will go down as far as constraints allow, but if it reaches the barrier 
> > or zero it will reject the trade
> > and all lower-ranked signals.
> > Setting signal's pos size to zero in the beginning effectively means the 
> > same condition.
> > 
> > You should NOT set position size to zero if you want to reject single trade 
> > BUT continue to handle lower-ranked signals.
> > 
> > If you want to SKIP one signal, without affecting others, you should set 
> > Price property of that signal to -1 (minus one).
> > 
> > Best regards,
> > Tomasz Janeczko
> > amibroker.com
> > 
> > On 2010-07-20 12:42, rise_t575 wrote:
> > >
> > > Thanks a lot - I will try your solution.
> > > Do you have an idea *why* this is happening?
> > >
> > > --- In [email protected], Tavan Taban<tavantaban@>  wrote:
> > >> I remember experiencing the same problem earlier. As far as I remember, 
> > >> it
> > >> is something like, if it cancels one, cancels also the rest which are not
> > >> coded to be rejected. I can dig more if it helps.
> > >>
> > >> Anyway, one solution alternative may be the following.
> > >>
> > >> for( sig = bo.GetFirstSignal( bar ); sig; sig = bo.GetNextSignal( bar ) )
> > >>          {
> > >>              if (IDontLikeThisSignal) sig.Type = 7;
> > >>          } // end if exit
> > >>
> > >>
> > >> 2010/7/17 rise_t575<rise_t@>
> > >>
> > >>>
> > >>> Hello,
> > >>>
> > >>> I've noticed using mid-level CBT that when I set the position size to 
> > >>> zero
> > >>> for the signal in question (the reason for setting it to zero is 
> > >>> slightly
> > >>> complicated&  not that important here - some data needed for a 
> > >>> subsequent
> > >>> calculation is {empty}), the trade is marked as "rejected" in AA's 
> > >>> results
> > >>> list (which is perfectly ok).
> > >>> What is not "perfectly ok" is the fact that the backtester rejects the
> > >>> following signals at the same bar as well (there's enough cash 
> > >>> available and
> > >>> position size is>  0).
> > >>>
> > >>> How can I prevent this?
> > >>>
> > >>>
> > >>>
> > >
> > >
> > >
> > > ------------------------------------
> > >
> > > **** IMPORTANT PLEASE READ ****
> > > This group is for the discussion between users only.
> > > This is *NOT* technical support channel.
> > >
> > > TO GET TECHNICAL SUPPORT send an e-mail directly to
> > > SUPPORT {at} amibroker.com
> > >
> > > TO SUBMIT SUGGESTIONS please use FEEDBACK CENTER at
> > > http://www.amibroker.com/feedback/
> > > (submissions sent via other channels won't be considered)
> > >
> > > For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
> > > http://www.amibroker.com/devlog/
> > >
> > > Yahoo! Groups Links
> > >
> > >
> > >
> > >
> >
>


Reply via email to