[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [EquisMetaStock Group] Re: EST long bias problem



PureBytes Links

Trading Reference Links

That is my problem!!

I need something like the Alert function in your #2 choice to test what the previous position was. A normal Alert will not do as the "trigger" for the previous trade usually have returned to <1, which means that both the long and short triggers might be <1 just before there is a new trigger. 

I need something like "IsLong" or "IsShort" to test when a new trigger appears. Some variable needs to be switched to Long/Short and needs to keep that value until the next signal. When the new signal appears, it must then test this variable to decide if a new trade in the opposite direction can be made.

Carl :(

  ----- Original Message ----- 
  From: Jose Silva 
  To: equismetastock@xxxxxxxxxxxxxxx 
  Sent: Monday, November 13, 2006 10:14 PM
  Subject: [EquisMetaStock Group] Re: EST long bias problem


  Ok, and...

  4)If Entry & Exit signals appear on the same bar and the previous 
  signal was *both long & short*, then give preference to ...

  It would be so much simpler if the buy & sell conditions didn't clash.

  jose '-)
  http://www.metastocktools.com

  --- In equismetastock@xxxxxxxxxxxxxxx, "Carl van Heerden" <cj.vh@xxx> 
  wrote:
  >
  > Jose.....triple Dutch to me!! I feel like a tulip whacked by a 
  windmill....
  > 
  > Let me try again:
  > The Entry signal is when the Short term RWIL spikes through 1 while
  > the Long term RWIH is above 1.
  > The Exit signal is when the Short term RWIH spikes trough 1 while
  > the Long term RWIL is above 1
  > 
  > 1) If Entry & Exit appears on the same bar and it is the FIRST
  > signal, the trade can be ignored for this purpose. It has not
  > happened yet in the different data sets I have tested (i.e. the
  > first trade has not been Entry & Exit on the same bar), but it can
  > conceivably happen.
  > 
  > 2)If Entry & Exit signals appear on the same bar and the previous
  > signal was long, then give preference to the short signal.
  > 
  > 3)If Entry & Exit signals appear on the same bar and the previous
  > signal was short, then give preference to the long signal.
  > 
  > Your #2 option does not signal a trade when the two signals appear
  > on one bar.
  > 
  > ..........hope this is clearer. I appreciate your help.
  > 
  > Carl
  > 
  > 
  > ----- Original Message ----- 
  > From: Jose Silva 
  > To: equismetastock@xxxxxxxxxxxxxxx 
  > Sent: Sunday, November 12, 2006 8:51 PM
  > Subject: [EquisMetaStock Group] Re: EST long bias problem
  > 
  > 
  > Carl, I'm having trouble understanding your request - the 
  description 
  > reads like double dutch to me. :) (Apologies for the pun)
  > 
  > The most difficult part of programming any strategy, is 
  understanding 
  > and getting the idea across to others. A strategy may at first 
  seem 
  > easy to understand from its creator's point of view, but 
  translating 
  > it into understandable terms for others is quite an art form.
  > 
  > Try explaining the problem in logical terms:
  > 
  > 1) If Entry & Exit signals appear on the same bar, give 
  preference to
  > ... signal;
  > 
  > 2) If Entry & Exit signals appear on the same bar, and previous 
  bar 
  > was Long, give preference to ... signal;
  > 
  > 3) If Entry & Exit signals appear on the same bar, and previous 
  bar 
  > was Short, give preference to ... signal;
  > 
  > ... And so on.
  > 
  > jose '-)
  > http://www.metastocktools.com
  > 
  > --- In equismetastock@xxxxxxxxxxxxxxx, "Carl van 
  > Heerden" <cj.vh@> wrote:
  > >
  > > Hi Jose,
  > > 
  > > I appreciate your reply.
  > > 
  > > The difference between the MA cross-over used in your example 
  and 
  > the RWI is that the RWI signals the start of the trade with a one 
  > bar "spike" through 1. The indicator might stay above 1 for 
  several 
  > bars or might return to below one immediately, but the signal 
  will 
  > still be valid (unlike with a MA where the signal reverses when C 
  > crosses the MA). The reverse RWI signal (say the SHORT signal) 
  > takes place when a slightly different indicator spikes through 1. 
  > This might even happen while the "Long" indicator is still above 
  1. 
  > This would still indicate a reversal. I want the system to then 
  take 
  > the opposite position, even though both signals are above one. 
  The 
  > Alert function is difficult to use in this case.
  > > 
  > > Regards
  > > 
  > > Carl
  > > 
  > > 
  > > 
  > > 
  > > ----- Original Message ----- 
  > >
  > > From: Jose Silva 
  > > To: equismetastock@xxxxxxxxxxxxxxx 
  > > Sent: Saturday, November 11, 2006 10:56 AM
  > > Subject: [EquisMetaStock Group] Re: AST long bias problem
  > > 
  > > Carl, take a look at the indicator code below, and in particular
  > > note how option #2 deals with same-bar trades.
  > > 
  > > MetaStock -> Tools -> Indicator Builder -> New ->
  > > -> copy & paste complete formulae between "---8<---" lines.
  > > 
  > > ====================
  > > System trade signals
  > > ====================
  > > ---8<---------------------------
  > > 
  > > { System trade signals v3.2
  > > 
  > > ©Copyright 2005-2006 Jose Silva
  > > For personal use only.
  > > http://www.metastocktools.com
  > > 
  > > Note on simultaneous entry/exit on same bar:
  > > #[1] Ignore:
  > > entry/exit signals cancel each other;
  > > #[2] Preference (given to):
  > > new entry if currently Short,
  > > new exit if currently Long.}
  > > 
  > > { Signals reference example }
  > > entry:=C>Mov(C,21,E);
  > > exit:=C<Mov(C,10,E);
  > > 
  > > { User inputs }
  > > plot:=Input("Signals: [1]Clean, [2]All, [3]Trade binary",1,3,1);
  > > choose:=Input("Simultaneous entry/exit: [1]Ignore, 
  > > [2]Preference",1,2,2);
  > > delay:=Input("Entry and Exit delay",0,5,0);
  > > 
  > > { Initialize signals }
  > > init:=Cum(IsDefined(entry+exit))=1;
  > > 
  > > { #1 - Ignore entry/exit on same bar }
  > > bin1:=ValueWhen(1,entry-exit<>0 OR init,entry);
  > > long:=bin1*(Alert(bin1=0,2)
  > > OR entry*Cum(entry)=1);
  > > short:=(bin1=0)*(Alert(bin1,2)
  > > OR exit*Cum(exit)=1);
  > > signals1:=long-short;
  > > 
  > > { #2 - Preference to first entry/exit }
  > > long:=entry*(Alert(entry=0,2)
  > > OR entry*Cum(entry)=1);
  > > short:=exit*(Alert(exit=0,2)
  > > OR exit*Cum(exit)=1);
  > > bin2:=ValueWhen(1,long-short<>0 OR init,long);
  > > long:=bin2*(Alert(bin2=0,2)
  > > OR entry*Cum(entry)=1);
  > > short:=(bin2=0)*(Alert(bin2,2)
  > > OR exit*Cum(exit)=1);
  > > signals2:=long-short;
  > > 
  > > { Select #1 or #2 }
  > > binary:=If(choose=1,bin1,bin2);
  > > signals:=If(choose=1,signals1,signals2);
  > > 
  > > { Plot in own window }
  > > Ref(If(plot=2,entry,0),-delay);
  > > Ref(If(plot=1,signals,
  > > If(plot=2,-exit,binary)),-delay)
  > > 
  > > ---8<---------------------------
  > > 
  > > jose '-)
  > > http://www.metastocktools.com
  > >
  > >
  > > 
  > > --- In equismetastock@xxxxxxxxxxxxxxx, "Carl van 
  Heerden" <cj.vh@> 
  > > wrote:
  > >
  > > Hi Jose,
  > > 
  > > Here is the basic code
  > > 
  > > To go long:
  > > RWIL(opt1,opt2)>1 
  > > 
  > > AND RWIH(opt3,opt4)>1 
  > > 
  > > To go Short:
  > > RWIH(opt1,opt2)>1 
  > > 
  > > AND RWIL(opt3,opt4)>1
  > > 
  > > Straight forward, until it happens that both triggers are on the
  > > same bar. Then the Order Bias is forced either Long or Short
  > > according to your choice in EST. As I wrote before, the trades 
  need
  > > to alternate.
  > > 
  > > Regards
  > > 
  > > Carl



   

  __________ NOD32 1863 (20061113) Information __________

  This message was checked by NOD32 antivirus system.
  http://www.eset.com


[Non-text portions of this message have been removed]




 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/equismetastock/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/equismetastock/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:equismetastock-digest@xxxxxxxxxxxxxxx 
    mailto:equismetastock-fullfeatured@xxxxxxxxxxxxxxx

<*> To unsubscribe from this group, send an email to:
    equismetastock-unsubscribe@xxxxxxxxxxxxxxx

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/