I have been using AMTX when I run informal tests (all my tests are informal, it seems).
I've tested the first part of BradCCI (the one with the parentheses in the traditional place) with this system:
Enter long: mov(fml("BradCCI Line 1"), 5, s) >mov(fml("BradCCI Line 1"), 34, s)
Exit long: mov(fml("BradCCI Line 1"), 5, s) <mov(fml("BradCCI Line 1"), 34, s)
BradCCI Line 1: (((H+L+C)/3)-Mov(C,28,S))/(.015*Std(C,28))
Oddly enough, if I do the same test with BradCCI without the parentheses in the divisor, I get better results in profit/system testing. Amusing, huh?
Both seem to do a good job catchting the signficant highs and lows, particularly the jump in the spring of '96, but both give too many signals the rest of the time. The system without the parentheses gives slightly fewer signals.
All this is just chitchat, I know -- not a report of extensive tests. I don't know whether it means much. I confess I'm suspicious of a system based on the one without parentheses, because there's no rationale for dividing the dividend by .015. Well, there may be, but it hasn't been stated yet.
The number of signals can be reduced by adding the second part of BradCCI, Slow Stochastics and a Price Oscillator. But still, it doesn't reduce them enough.
That system is:
Enter long:
mov(fml("BradCCI Line 1"), 5, s) >mov(fml("BradCCI Line 1"), 34, s) and
mov(stoch(8,5),5,s)>mov(stoch(8,5),34,s)
and mov(fml("BradCCI Line 2"), 5, s) >mov(fml("BradCCI Line 2"), 34, s)
{and mov(OscP(13,55,E,% ), 5, s) >mov(OscP(13,55,E,% ), 34, s)}
Enter long:
mov(fml("BradCCI Line 1"), 5, s) <mov(fml("BradCCI Line 1"), 34, s) and
mov(stoch(8,5),5,s)<mov(stoch(8,5),34,s)
and mov(fml("BradCCI Line 2"), 5, s) <mov(fml("BradCCI Line 2"), 34, s)
and mov(OscP(13,55,E,% ), 5, s) <mov(OscP(13,55,E,% ), 34, s)
I went looking for a book on CCI today, to get to the root of the .015 and its placement, after tossing and turning all night with various equations. The hilarious thing is that once I got to the store, I couldn't get out of the car because of a thunder and lightning storm that flooded the parking lot. The power was knocked out, too. Some heavenly diabolical trickster is making merry with BNS.
Really, the .015 is there to make a plot with a scale that moves basically between -100 and 100. You get a very odd scale when you change the purpose of .015.
I shoulda stuck with Bollinger bands.
|
