No. of Recommendations: 4
<<How y'all doing. >>

The last two months have been a nightmare. But thanks for asking. :)

<<Nice to see that you haven't lost that contrarian streak. >>

Not me. I'm a trend follower. Seeing as the trends reverse themselves the instant they start nowadays, I have lost money hand over fist the last two months. But I'm still in there swinging (and missing.)

<<As always - you can be relied on to provide another opinion - it makes the rest of us sheep stop to think. A valuable contribution. Peter Eliades thought his research had produced 'remarkable' signs of a reliable pattern. Six out of Seven and one near miss in 73 years, is worth paying attention to >>

Six out of seven?

Well, if the sign was a meaningless indicator, it would then work half the time - which would mean 3 or 4 times out of 7.

Does 6 out of 7 represent a level of statistical significance (p<.05) above the chance level (3.5 out of 7.) I do not believe it does. But even that is not the best way to test it - 7 out of 7 would be statistically significant but I would still it doesn't have to mean anything.

<<Since I am asking you for a forecast let me give you mine:
we are in a countertrend rally(I won't refer to the Elliott Count because you don't think much of that approach either) the nominal 49 day cycle has stretched to 51 days and is due to touch down on 10/12/00 (+/- 1 day).>>

I don't make forecasts. I do not believe it is possible for anyone to predict the direction of the market. And as relentlessly wrong as I've been the last two months solid, any prediction I would make would simply be something to fade.

However, the inherent problem here is the same one with backtesting and data mining.

Just as ELiades has found the "Sign of the Bear" anyone else could just as easily set their algorithm to search for other patterns which "predicted" bull runs/bear runs or whatever.

We could find thousands of such patterns depending on what parameters we search for. This doesn't give them any predictive value.

If we could optimize our trading systems based on back testing, we'd all be millionaires. Either that or the market would be 100% efficient. But we can't.

Going back in time and finding that Point B was preceded by Point A, even when every Point A was followed by Point B, in no way suggests the next instance of Point A will be followed by Point B.

That's just simple statistics.

Print the post  


When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and Glassdoor #1 Company to Work For 2015! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.