PureBytes Links
Trading Reference Links
|
The respective AA research for the single-level Long MeanD-ratio
system is
D5=Foreign("~Dratio5","C");
D10=Foreign("~Dratio10","C");
D15=Foreign("~Dratio15","C");
D20=Foreign("~Dratio20","C");
D25=Foreign("~Dratio25","C");
D30=Foreign("~Dratio30","C");
D35=Foreign("~Dratio35","C");
D40=Foreign("~Dratio40","C");
K=Optimize("K",10,5,40,5);
D=IIf(K==5,D5,IIf(K==10,D10,IIf(K==15,D15,
IIf(K==20,D20,IIf(K==25,D25,IIf(K==30,D30,IIf(K==35,D35,D40)))))));
X=101*Optimize("X",20,7,37,1);// my database has 101 symbols, 100+^NDX
Buy=Cross(D,X);Sell=Cross(X,D);
// [10,20]=+91%%
BTW, this [10,20] solution is great for some reasons :
a. Only 4 non-profitable stocks [-13%, -11%, -4%, -0.10%] and the
rest 97
are profitable up to +389%.
b. The index itself has the same optimum and goes to +75%
c. The contrarian system Sell=Cross(D,X);Buy=Cross(X,D); is *very*
miserable, -87% for the index and
-50% for the market, which makes me sure for the results of
the ...wrong choice. Indeed, at some "buy"
signals the market was severely selling, you should have a lot of
discipline to trust the "correct" signal.
d. An almost double price in this bearish environment is very good
for a Long system.
e. The exposure is only 13%, 18 quick trades since Jan2000
f. It was offering a great support the cloudy period Oct2000-
Sept2001. You remember a lot of traders have had
great damages this period, until they understood the bearish
character of the market.
g. It is charmy the last two years : It gives a high quality Buy hint
[Sept2001, endJuly2002] and mainly stays
out of the market, waiting for ...better D-ratio days. I really like
it, take a look !!
Dimitris Tsokakis
--- In amibroker@xxxxxxxxxxxxxxx, "DIMITRIS TSOKAKIS" <TSOKAKIS@xxxx>
wrote:
> Rick,
> when I searched the MeanD-ratio, for example, I just created 10
> composite tickers
> Dratio=500*(H-L)/(H+L);
> Dratio5=DEMA(Dratio,5);
> Dratio10=DEMA(Dratio,10);
> Dratio15=DEMA(Dratio,15);
> Dratio20=DEMA(Dratio,20);
> Dratio25=DEMA(Dratio,25);
> Dratio30=DEMA(Dratio,30);
> Dratio35=DEMA(Dratio,35);
> Dratio40=DEMA(Dratio,40);
> AddToComposite(Dratio5,"~Dratio5","C");
> AddToComposite(Dratio10,"~Dratio10","C");
> AddToComposite(Dratio15,"~Dratio15","C");
> AddToComposite(Dratio20,"~Dratio20","C");
> AddToComposite(Dratio25,"~Dratio25","C");
> AddToComposite(Dratio30,"~Dratio30","C");
> AddToComposite(Dratio35,"~Dratio35","C");
> AddToComposite(Dratio40,"~Dratio40","C");
> Buy=0;
> and then watched the trading behavior of the new ~tickers.
> From experience, the smoothing period of the initial D-ratio
depends
> on the trading system that follows : A one-level MeanD-ratio
prefers
> faster smoothing, a two-level was better with slower ones.
> Dimitris Tsokakis
> --- In amibroker@xxxxxxxxxxxxxxx, "Rick Parsons" <RickParsons@xxxx>
> wrote:
> > I have a system similar to Dimitri's Trade the Market where I use
> AddToComposite to count the number of stocks above or below the
MACD
> Signal line.
> >
> > Question is, what is the best MACD period to use? It is very
> cumbersome to change the periods, run the Scan to create an
> AddToComposite, then backtest it.
> >
> > Does anyone have a creative idea on how to "optimize" the periods
> used to create the AddToComposite?
> >
> > Thanks,
> > Rick
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Rent DVDs from home.
Over 14,500 titles. Free Shipping
& No Late Fees. Try Netflix for FREE!
http://us.click.yahoo.com/BVVfoB/hP.FAA/uetFAA/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/
|