PureBytes Links
Trading Reference Links
|
Dimitris,
The linear (Least-Squares ) regression model
postulates that
Y= a+bX+e
Where the "residual" e is a random variable with mean zero.
The coefficients a and b are determined by the
condition that the sum of the square residuals is as small as possible.
So error is SUM OF THE SQUARE of residuals.
You are not using sum of squares of residuals as error estimate.
For more info see:
<A
href="">http://www.stat.yale.edu/Courses/1997-98/101/linreg.htm
Best regards,Tomasz Janeczkoamibroker.com
<BLOCKQUOTE
>
----- Original Message -----
<DIV
>From:
Dimitris
Tsokakis
To: <A title=amibroker@xxxxxxxxxxxxxxx
href="">amibroker@xxxxxxxxxxxxxxx
Sent: Monday, March 15, 2004 12:17
PM
Subject: [amibroker] The linear trend
calculations
I tried to duplicate the LinRegIntercept( ARRAY, periods )
example from the 4.50 users guide.I used as example the
C1=MA(RSI(50),200).The error was calculated from
Cum(abs(y-C1)).
C1=MA(RSI(50),200);Plot(C1,"C1",1,1); x =
Cum(1);lastx = LastValue( x ); Daysback =
LASTX-ValueWhen(Cum(IsTrue(C1))==1,X); aa = LastValue( LinRegIntercept(
C1, Daysback) );bb = LastValue( LinRegSlope( C1, Daysback ) );y = Aa +
bb * ( x - (Lastx - DaysBack) ); Plot( IIf( x >= (lastx - Daysback), y,
-1e10 ), "LinReg", colorRed,8 );
Title=Name()+"\nLine="+WriteVal(aa,1.1)+"+"+WriteVal(bb,1.3)+"*(t-t1)"+"\nError="+WriteVal(Cum(abs(y-C1)),1.0);start=Cum(IsTrue(C1))==1;PlotShapes(shapeUpTriangle*start,colorWhite);
My analytic
codet=Cum(1)-1;tmax=LastValue(t);C1=MA(RSI(50),200);Plot(C1,"C1",1,1);C1max=1.5*LastValue(ceil(Highest(C1)));C1min=0.5*LastValue(floor(Lowest(C1)));start=Cum(IsTrue(C1))==1;t1=ValueWhen(start,t);PlotShapes(shapeUpTriangle*start,colorWhite);C10=ValueWhen(start,C1);d0=Param("d0",-0.03,-0.03,-0.005,0.005);Line0=C10+d0*(t-t1);d1=-d0;Line1=C10+d1*(t-t1);Error0=LastValue(Cum(abs(C1-Line0)));Error1=LastValue(Cum(abs(C1-Line1)));Err=Max(Error0,Error1);Title=Name()+"\nInitial
Error="+WriteVal(err,1.0);stepi=0.1;stepj=0.001;j1=d1;j0=d0;for(i=C1min;i<=C1max;i=i+stepi){for(j=j0;j<=j1;j=j+stepj){Line=i+j*(t-t1);err1=LastValue(Cum(abs(C1-Line)));if(err1<err)err=Min(err,err1);}}Title=Title+"\nMinError="+WriteVal(err,1.0);iA=0;jA=0;for(i=C1min;i<=C1max;i=i+stepi){for(j=j0;j<=j1;j=j+stepj){Line=i+j*(t-t1);err1=LastValue(Cum(abs(C1-Line)));if(err1==err){Plot(Line,"\n"+WriteVal(j,1.4),colorRed,8);Title=Title+"\n\nLine="+WriteVal(i,1.1)+"+"+WriteVal(j,1.3)+"*(t-t1)";iA=i;jA=j;}}}Title=Title+"\nFirst
Approx [iA,jA]=["+WriteVal(iA,1.1)+","+WriteVal(jA,1.3)+"]";
gave, in many cases, smaller error.Is there any
explanation ?Dimitris Tsokakis
Send BUG REPORTS to
bugs@xxxxxxxxxxxxxSend SUGGESTIONS to
suggest@xxxxxxxxxxxxx-----------------------------------------Post
AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: <A
href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check
group FAQ at: <A
href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html
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
Yahoo! Groups Sponsor
ADVERTISEMENT
Yahoo! Groups Links
To visit your group on the web, go to:http://groups.yahoo.com/group/amibroker/
To unsubscribe from this group, send an email to:amibroker-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
|