|
PureBytes Links
Trading Reference Links
|
And, finally, instead all this pages [or vols] about robustness, walk-
forward testing, safe optimization timeframes [3 years before and 3
years after, as if we could say the same things back in 1990 !!!],
the tough ^NDX [^IXIC] market was nothing but one click [or
better...two].
Try the ultra-simple
// ^NDX, ^IXIC, 1/1/1997 till now
x=IIf(DateNum()<1000103 OR DateNum()>1021001,20,60);y=IIf(x==20,19,5);
Buy=Cum(1)%x==0;
Sell=Ref(Buy,-y);
Short=Sell;Cover=Buy;
and you will probably agree...
Dimitris Tsokakis
[expecting, not so soon, the next click]
--- In amibroker@xxxxxxxxxxxxxxx, "DIMITRIS TSOKAKIS" <TSOKAKIS@xxxx>
wrote:
> Dave,
> I think the analytic STORSI example just posted answers many of
your
> questions.
> The system was an example how to use STORSI, it was not the best
> STORSI system, it was not the best QQQ system and, finally, it was
> not a trading suggestion.
>
> --- In amibroker@xxxxxxxxxxxxxxx, "Dave Merrill" <dmerrill@xxxx>
> wrote:
> > One further related question: You mention that this is "tuned to
> the bearish
> > environment". Do you have an indicator that helps you know
> quantitatively
> > when we're no longer in a bearish environment?
>
> there is an interesting answer [among many others...]. Take a look
at
> the equity curve of this system. It is a very good trend indicator.
> It goes up, when the market goes down and refuses to follow the
> bullish mood of the last year or so !!
> This is an interesting method for me.
> I have designed some systems and watch their performance. They were
> very interesting until Sept2002 and, since then, they have a slow
but
> constant loss rate. The first impression was ah ! this system is
not
> good, it is not an all-time blind profit machine, it will never be
a
> grail etc, etc. We often read similar comments in this list and I
> really wonder if these comments are serious or not. Give a second
> chance, take another closer look and you will probably see that
> a "non-robust" system Equity may be an interesting trend indicator.
> The more "non-robust" the better. In this way, you may find out a
> really contrarian system, the pros here will post quick and sharp
> comments about the validity of the system, but, they always sing
the
> same song, as if they press every member to post here an all-time
> profitable system [what for ???] and it is better to look [and take
> care] what you´ve got, except if you don´t agree that the
systematic
> anti-holy-grail is much better than the [non-existing] holy grail.
> This STORSI variation [and not only this] have this charm to show
you
> the market changes through their [non-robust] Equity curve.
> I hope it will throw a light in this untrodden area...
> Dimitris Tsokakis
> > Thanks,
> >
> > dave
> > Dimitris, thanks for the ideas, I'll check it out. Besides a
QQQ
> trading
> > vehicle, I'm interested in the smoothing method you chose, and
also
> your
> > trend qualifier.
> >
> > Just out of curiosity, what commission settings do you use? I'm
> at 1%,
> > which is my usual, but if that's too stringent in this context,
> maybe that's
> > why my initial results with this, and other things I've tried,
> aren't as
> > spectacular.
> >
> > I know some of what I'm about to ask has been discussed here in
> general,
> > but let me ask it in this specific context anyway: There are a
> number of
> > parameters here, either adjustable or built into the code. If we
> wanted to
> > optimize for better performance in the current environment, what
> test period
> > would you think would translate well to the future? Going
forward,
> when
> > would you think you'd want to optimize again? How would you
suggest
> we
> > distinguish a beautiful curve fit over the test period from
> settings likely
> > to perform well in the future?
> >
> > Please don't take those questions as anything other than what
> they are, a
> > sincere desire to learn.
> >
> > Dave
> > Dave,
> > It would be interesting to make your own study on the STORSI.
> > Here is an example for QQQ.
> > It was tuned to the bearish environement, the peak of the
> Equity is
> > very close to the change of the trend.
> >
> > // A STORSI example for QQQ
> > SetBarsRequired(1000,1000);
> > function IIR2( input, f0, f1, f2 )
> > {
> > result[ 0 ] = input[ 0 ];result[ 1 ] = input[ 1 ];
> > for( i = 100; i < BarCount; i++ )
> > {
> > result[ i ] = f0 * input[ i ] + f1 * result[ i - 1 ] + f2
*
> result
> > [ i - 2 ];
> > }
> > return result;
> > }
> > x=DEMA(RSI(10),10);t=20;
> > C1=100*(x-LLV(x,t))/(HHV(x,t)-LLV(x,t));Plot(C1,"STORSI",1,1);
> > K=0.4;
> > RD=IIR2( C1, 0.3, 1.6, -0.9);// SMOOTHED STOCHRSI
> > Plot(RD,"["+WriteVal(10*K,1.0)+"]",4+10*K,1);
> > TREND=LinearReg(DEMA(C,20),30);// TREND QUALIFIER
> > Plot(TREND,"TREND",4,styleOwnScale);
> > BLEVEL=Optimize("BL",-19,-30,-10,1);Plot(BLEVEL,"BLEVEL",1,1);
> > SLEVEL=Optimize("SL",49,40,60,1);Plot(SLEVEL,"SLEVEL",1,1);
> > Buy=Cross(RD,BLEVEL);Sell=Cross(RD,SLEVEL);
> > Short=Cross(RD,SLEVEL) ;Cover=Buy;
> >
> > It is a good point to start and tune it to the present
> conditions,
> > add trend characteristics [I have added a TREND function, you
> may
> > select yours] etc.
> > From June2000 till now, no losing combinations, affordable
> drawdowns
> > [less than 40%] and very good behavior when the bears define
the
> > game...
> > Better than +350%, winners/losers ~2/1, not bad at all for
QQQ.
> > Of course it is not the only QQQ vehicle, it is just an
example
> to
> > begin with.
> > Dimitris Tsokakis
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/GHeqlB/TM
---------------------------------------------------------------------~->
Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|