There are also a couple of errors in my original code:
- SetTradeDelays should be (1,1,1,1)
- MAFilter = C > MA(C,200) should be MAFilter = Foreign("^GSPC","C") >
MA(Foreign("^GSPC","C"),200);
The final point was that I found this code did not work for anyone using TC2005
data but worked fine with Yahoo data. That suggests that splits and/or
dividends affect the results. Given that splits are less common in these ETFs
(at least as far as I know - someone correct me if I'm wrong), I think it is a
dividend adjustment issue.
That last point indicates to me that using Telechart data for backtesting is
not viable.
Thoughts appreciated.
--- In [email protected], "donald_brown_48367" <donald_brown_48...@...>
wrote:
>
>
> 1. Copied put the SPDRs into a watchlist.
>
> 2. I copied the Droskill's code into the formula editor and saved it.
>
> 3. I ran a backtest filtering it just for the SPDR watchlist with the a
> start date of July 1, 1999 and an end date of Dec 23 2009.
>
>
> --- In [email protected], Bob Waits <bobwaits2@> wrote:
> >
> > Thanks Donald. Could you walk us through (steps) so that we can
> understand how you got these returns?
> >
> >
> >
> >
> > ________________________________
> > From: donald_brown_48367 donald_brown_48367@
> > To: [email protected]
> > Sent: Sat, January 2, 2010 12:15:07 PM
> > Subject: [amibroker] Re: CXO article
> >
> >
> > As I understand it your code would be duplicating the red line on the
> CXO chart. I used your code with a start date of July 1, 1999 and an end
> date of Dec 23, 2009 and I got a 53.9% profit over the whole period
> which is close to (slightly higher) than the red line on their chart as
> this increases from $10,000 to almost $15,000 over the same period. I
> used a .25% trading cost.
> >
> > The results aren't exactly the same for some reason. My results show a
> balance drawdown from April 2001 until March 2004. The red line on the
> chart appears to be underwater only from July 2002 until September 2003.
> >
> > --- In amibro...@yahoogrou ps.com, "droskill" droskill@ .> wrote:
> > >
> > > I've been trying to duplicate the results in this CXO article:
> > >
> > > http://www.cxoadvisory.com/blog/internal/blog12-22-09/
> > >
> > > It's a rotational system that trades the top fund from the SPDR
> Select series based on a 6 month return as long as it is above the 10
> month SMA.
> > >
> > > I'm not testing it monthly, but here's the daily code I've come up
> with:
> > >
> > > SetBacktestMode( backtestRotation al);
> > > SetTradeDelays( 0,0,0,0); // everything delayed 1 day
> > > SetOption("UsePrevB arEquityForPosSi zing", True);
> > > SetOption("MinShare s", 10);
> > > SetOption("AllowPos itionShrinking" ,True);
> > > SetOption("AccountM argin",100) ;
> > >
> > > Totalpositions = 1;
> > >
> > > MAFilter = C > MA(C,200);
> > > PositionScore = MAFilter * ROC(C,120);
> > > PositionSize = -100/Totalpositions ;
> > >
> > > SetOption("WorstRan kHeld", Totalpositions + 1);
> > > SetOption("MaxOpenP ositions" , Totalpositions );
> > >
> > > m = Month();
> > > newMonth = m != Ref( m, -1);
> > > PositionScore = IIf(PositionScore < 0, 0, PositionScore) ; // Long
> only
> > > PositionScore = IIf(newMonth, PositionScore, scoreNoRotate) ;
> > >
> > > My portfolio consists of the nine SPDRs - XLF,XLE,XLP, XLY,XLI,XLK,
> XLV,XLB,XLU
> > >
> > > I've been unable to duplicate anything close to their results -
> anybody have any idea of what, if anything, I'm doing wrong?
> > >
> >
>
