|
PureBytes Links
Trading Reference Links
|
Peter,
thank you very much for your reply.
I can not make a lot of comments on your code, see my
http://groups.yahoo.com/group/amibroker/message/39501
experimental idea.
I hope you will excuse my naive comments below.
Dimitris Tsokakis
--- In amibroker@xxxxxxxxxxxxxxx, "bluesinvestor" <investor@xxxx>
wrote:
> Dimitris,
>
> It does not seem to make a big difference with a variable IAF.
>
> Peter
>
> // Initialize IAF Array
> X=(DEMA(StochD(40),20)-50)/50;
> IAF=0.05*(1.1+abs(X));
this function varies [for ^NDX] from 0.055 to 0.105
It is important for the MaxAF selection
>
> function SARafl( Cvar, Hvar, Lvar, MaxAF )
> {
> //IAF = 0.02; // acceleration factor
> //MaxAF = 0.02; // max acceleration
It does not make sense to me. MaxAF should be at least 10*IAF
> psar = Cvar; // initialize
> long = 1; // assume long for initial conditions
> af = IAF[ 0 ]; // init acelleration factor
I suppose here you define the initial af value
> ep = Lvar[ 0 ]; // init extreme point
> hp = Hvar [ 0 ];
> lp = Lvar [ 0 ];
>
> for( i = 2; i < BarCount; i++ )
> {
> if ( long )
> {
> psar [ i ] = psar [ i-1 ] + af * ( hp - psar [
> i-1 ] );
I suppose here you create the various psar values.
af has still it initial value ? I do not see variable af here.
> }
> else
> {
> psar [ i ] = psar [ i-1 ] + af * ( lp - psar [
> i-1 ] );
af still has its initial value ?
> }
>
> reverse = 0;
> //check for reversal
> if ( long )
> {
> if ( Lvar [ i ] < psar [ i ] )
> {
> long = 0; reverse = 1; // reverse
> position to Short
> psar [ i ] = hp; // SAR is Hvar
> point in prev trade
> lp = Lvar [ i ];
> //af = IAF;
> af = IAF[ i ];
> }
> }
> else
> {
> if ( Hvar [ i ] > psar [ i ] )
> {
> long = 1; reverse =
1; //reverse
> position to long
> psar [ i ] = lp;
> hp = Hvar [ i ];
> //af = IAF;
> af = IAF[ i ];
> }
> }
>
> if ( reverse == 0 )
> {
> if ( long )
> {
> if ( Hvar [ i ] > hp )
> {
> hp = Hvar [ i ];
> //af = af + IAF;
> af = af + IAF[ i ];
> if( af > MaxAF ) af = MaxAF;
> }
>
> if( Lvar[ i - 1 ] < psar[ i ] ) psar[
i
> ] = Lvar[ i - 1 ];
> if( Lvar[ i - 2 ] < psar[ i ] ) psar[
i
> ] = Lvar[ i - 2 ];
> }
> else
> {
> if ( Lvar [ i ] < lp )
> {
> lp = Lvar [ i ];
> //af = af + IAF;
> af = af + IAF[ i ];
> if( af > MaxAF ) af = MaxAF;
> }
>
> if( Hvar[ i - 1 ] > psar[ i ] ) psar[
i
> ] = Hvar[ i - 1 ];
> if( Hvar[ i - 2 ] > psar[ i ] ) psar[
i
> ] = Hvar[ i - 2 ];
>
> }
> }
> }
> return psar;
> }
>
> Plot( Close, "Price", colorBlack, styleCandle );
> Plot( SARafl(C,H,L,0.02), "SAR", colorRed, styleDots | styleNoLine |
> styleThick );
>
> Filter=1;
> AddColumn(SAR(.02,.02),"SAR");
you use the same IAF, maxAF here ?
> AddColumn(SARafl(C,H,L,0.02),"SARafl");
0.02 is very low for maxIAF
> AddColumn(iaf,"IAF");
>
> -----Original Message-----
> From: DIMITRIS TSOKAKIS [mailto:TSOKAKIS@x...]
> Sent: Friday, May 02, 2003 2:57 PM
> To: amibroker@xxxxxxxxxxxxxxx
> Subject: [amibroker] Re: DIMITRIS- Question
>
> Peter,
> Tough question [I hope the same for the answer...].
> To use realistic examples, the
>
> X=(DEMA(StochD(40),20)-50)/50;// the divergence of the smoothed
> stochastics from the average line
> IAF=0.05*(1.1+X);Plot(IAF,"",4,1);
>
> can make IAF vary from 0.01 [bearish] to 0.1[bullish]
> For an alternative, the
>
> X=(DEMA(StochD(40),20)-50)/50;
> IAF=0.05*(1.1-X);Plot(IAF,"",4,1);
>
> can make IAF vary from 0.01 [bullish] to 0.1 [bearish].
> You can also arrange IAF to be fast at the end of the trend
[bullish
> OR bearish] and slow during the trend with the simple alternative
>
> X=(DEMA(StochD(40),20)-50)/50;
> IAF=0.05*(1.1+abs(X));Plot(IAF,"",4,1);
>
> this would keep you close to the main curve at decision points
> and far from the main curve when it is better to go with the trend
> and wait.
> This is an initial example to control the SAR exits according to
the
> market behavior and SAR may be *any* indicator companion.
> The last 3 years we saw fast uptrends and prolonged downtrends, an
> opposite to the 98-99 days.
> It does not make sense to use *always* the same IAF because you do
> not have functions to do the right thing.
> On the other side, the use of variable parameters is the next,
> advanced, T/A step.
> Bullish, congestive and bearish phase can be [and should always be]
> just numbers.
> [It is the basic Pythagorean principle, true the last 2500 years]
> A good trend detector may give you these numbers.
> A good T/A software should help you to introduce these numbers
> everywhere, get the refined results, apply them to the market and
> catch the bird.
> Without variable parameters we shall be permanent victims of the
> whipsaws [and I do not like it !!!]
> Thanks for the creative communication.
> Dimitris Tsokakis
>
> --- In amibroker@xxxxxxxxxxxxxxx, "bluesinvestor" <investor@xxxx>
> wrote:
> > Dimitris,
> >
> > IAF and AF are not variable in the code I provided, although they
> could
> > be. But what type of array(s) would be passed for that
> information? Or
> > would they be calculated on the fly?
> >
> > Regards,
> > Peter
> >
> > -----Original Message-----
> > From: DIMITRIS TSOKAKIS [mailto:TSOKAKIS@x...]
> > Sent: Friday, May 02, 2003 1:29 PM
> > To: amibroker@xxxxxxxxxxxxxxx
> > Subject: [amibroker] Re: DIMITRIS- Question
> >
> > Peter,
> > thank you for the effort.
> > The question for SAR is in variable IAF and/or AF.
> > This will make the initial SAR more flexible.
> > Is it possible through your code ?
> > As for the variable inputs, we can simply have a SAR of any
> indicator
> > with a simple H,L replacement :
> > Example [the MeanCMO and its SAR companion]
> >
> > m25=Foreign("~sumCMO25","C");// after creating the ~sumCMO25 in AA
> > m=Foreign("~count","v");
> > meanCMO25=m25/m;
> > DD=DEMA(MEANCMO25,15);
> > H=dd;L=dd;// the replacement
> > cc=SAR(0.01,0.2 );
> > This 0.01 works well for the N100 database, it would be better to
> > vary from 0.01 to 0.08 according to the trend evolution.
> > The same for 0.2, it would be better to go up to 0.8 sometimes.
> > Dimitris Tsokakis
> > --- In amibroker@xxxxxxxxxxxxxxx, "bluesinvestor" <investor@xxxx>
> > wrote:
> > > Hello Dimitris,
> > >
> > > A little modification from Tomasz original code and 'kitakste
> edho'
> > > (excuse my lousy Greek ;)
> > >
> > > Regards,
> > > Peter
> > >
> > > function SARafl( Cvar, Hvar, Lvar, IAF, MaxAF )
> > > {
> > > //IAF = 0.02; // acceleration factor
> > > //MaxAF = 0.02; // max acceleration
> > >
> > > psar = Cvar; // initialize
> > > long = 1; // assume long for initial conditions
> > > af = IAF; // init acelleration factor
> > > ep = Lvar[ 0 ]; // init extreme point
> > > hp = Hvar [ 0 ];
> > > lp = Lvar [ 0 ];
> > >
> > > for( i = 2; i < BarCount; i++ )
> > > {
> > > if ( long )
> > > {
> > > psar [ i ] = psar [ i-1 ] + af * ( hp - psar [
> > > i-1 ] );
> > > }
> > > else
> > > {
> > > psar [ i ] = psar [ i-1 ] + af * ( lp - psar [
> > > i-1 ] );
> > > }
> > >
> > > reverse = 0;
> > > //check for reversal
> > > if ( long )
> > > {
> > > if ( Lvar [ i ] < psar [ i ] )
> > > {
> > > long = 0; reverse = 1; // reverse
> > > position to Short
> > > psar [ i ] = hp; // SAR is Hvar
> > > point in prev trade
> > > lp = Lvar [ i ];
> > > af = IAF;
> > > }
> > > }
> > > else
> > > {
> > > if ( Hvar [ i ] > psar [ i ] )
> > > {
> > > long = 1; reverse =
> > 1; //reverse
> > > position to long
> > > psar [ i ] = lp;
> > > hp = Hvar [ i ];
> > > af = IAF;
> > > }
> > > }
> > >
> > > if ( reverse == 0 )
> > > {
> > > if ( long )
> > > {
> > > if ( Hvar [ i ] > hp )
> > > {
> > > hp = Hvar [ i ];
> > > af = af + IAF;
> > > if( af > MaxAF ) af = MaxAF;
> > > }
> > >
> > > if( Lvar[ i - 1 ] < psar[ i ] ) psar[
> > i
> > > ] = Lvar[ i - 1 ];
> > > if( Lvar[ i - 2 ] < psar[ i ] ) psar[
> > i
> > > ] = Lvar[ i - 2 ];
> > > }
> > > else
> > > {
> > > if ( Lvar [ i ] < lp )
> > > {
> > > lp = Lvar [ i ];
> > > af = af + IAF;
> > > if( af > MaxAF ) af = MaxAF;
> > > }
> > >
> > > if( Hvar[ i - 1 ] > psar[ i ] ) psar[
> > i
> > > ] = Hvar[ i - 1 ];
> > > if( Hvar[ i - 2 ] > psar[ i ] ) psar[
> > i
> > > ] = Hvar[ i - 2 ];
> > >
> > > }
> > > }
> > > }
> > > return psar;
> > > }
> > >
> > > Plot( Close, "Price", colorBlack, styleCandle );
> > > Plot( psar, "SAR", colorRed, styleDots | styleNoLine |
> styleThick );
> > >
> > > Filter=1;
> > > AddColumn(SAR(.02,.02),"SAR");
> > > AddColumn(SARafl(C,H,L,0.02,0.02),"SARafl");
> > >
> > > -----Original Message-----
> > > From: DIMITRIS TSOKAKIS [mailto:TSOKAKIS@x...]
> > > Sent: Friday, May 02, 2003 3:40 AM
> > > To: amibroker@xxxxxxxxxxxxxxx
> > > Subject: [amibroker] Re: DIMITRIS- Question
> > >
> > > Wally,
> > > You may go to
> > > http://groups.yahoo.com/group/amibroker/message/38370
> > > http://groups.yahoo.com/group/amibroker/message/38424
> > > for the two alternatives.[I hope enough explained].
> > > I try to introduce variable smoothing to an already useful
trend
> > > detector, the DEMA(StochD(40),20).
> > > It has nothing to do with PSAR, it is a stand-alone indicator.
> > > Since it is quite smooth, I use 3-bars peak to signal the
change
> of
> > > the direction.
> > > DT
> > > PS. We will follow a similar procedure, when PSAR will accept
> > > variable parameters, to improove PSAR results.
> > >
> > > --- In amibroker@xxxxxxxxxxxxxxx, "netbull2000"
> <netbull2000@xxxx>
> > > wrote:
> > > > I read in one of the messages here that you came up with some
> new
> > > > trend indicator that seems to be better than PSAR. Could you
> > please
> > > > point me to your relevant message concerning this indicator.
I
> am
> > > > interested in this as PSAR is about the only indicator I have
> > some
> > > > respect for.
> > > >
> > > > Thanks...
> > > > Wally
> > >
> > >
> > >
> > > Send BUG REPORTS to bugs@xxxx
> > > Send SUGGESTIONS to suggest@xxxx
> > > -----------------------------------------
> > > 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/
> >
> >
> >
> > Send BUG REPORTS to bugs@xxxx
> > Send SUGGESTIONS to suggest@xxxx
> > -----------------------------------------
> > 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/
>
>
>
> Send BUG REPORTS to bugs@xxxx
> Send SUGGESTIONS to suggest@xxxx
> -----------------------------------------
> 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/
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get A Free Psychic Reading!
Your Online Answer To Life's Important Questions.
http://us.click.yahoo.com/cjB9SD/od7FAA/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/
|