>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Worse, it would have worked better for about 6 weeks or more now, to sell when my indicator goes green and buy when it goes red. I recognized this almost immediately when this whipsawing started, so I am doing ok, but this is insane, and unusual.
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
If it makes you feel any better the same thing happened to me.
I've been trading Rydex 2xNaz & -2xNaz (Velocity & Venture) for over a year and a half using a propietary system that takes into account intra-day price and volume patterns of the QQQ and compares to past patterns to make a statistical prediction just before the close of each day.
It used to yield a predictive value of between 60% and 70% which I consider "pretty good". Now suddenly it stopped "working" on Jan 3rd 2003, (and it was THAT abrupt)and not only that, but it reversed itself, and now I'm getting 80% accuracy but by doing exactly the opposite. (Negative correlation)
This never happened before, and it really baffled me. I'm making money again as the "inverse" correlation is even better than before, but now I have no theorical explanation as why it works that way.
As my sample space is only the last 2 years (a nasty bear market) I assumed that something changed this January, and I'm speculating that might be that the market shifted into a budding bull , and that's why the results reversed themselves. (I doubt it, this doesn't look like a baby bull to me...) Or maybe now there's blatant intervention by the PPT... Who knows...
Anyway. I hope you get some solace from the fact that you are not alone, (it did help me) and let's keep on trying to extract some money from this market...
Huluriasquias
PS: BTW, I thing that whipsawing has nothing to do with it. My system used to work best in ranged markets, and no so good in trending ones.
AI
