Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Pastimes : Clown-Free Zone... sorry, no clowns allowed -- Ignore unavailable to you. Want to Upgrade?

To: pater tenebrarum who wrote (49502)	12/20/2000 7:49:30 AM
From: Lucretius	Read Replies (1) \| Respond to of 436258

sickso trading at 39 in europe.... got delta hedge ass f*cking? -g-

To: pater tenebrarum who wrote (49502)	12/20/2000 9:09:24 AM
From: Oblomov	Read Replies (2) \| Respond to of 436258

Heinz... I didn't mean to seem sarcastic. Although my math and stats experience has trained me to be skeptical about such concepts as time-segment analysis and Bradley cycles, at the same time I do not reject these methods. I have done extensive time series analysis on the 10yr bond yield (I trade the futures now and then), as well as some stocks that I trade both long and short. To be sure, there are tradable patterns that emerge. For example, using spectral analysis (a method of using Fourier transforms to find the frequency of pattern recurrence in a time series) and Cox regression (analysis of the length of "runs", or the distance from top-to-bottom, etc.) I was able to nail the January top and the May secondary top in the ^TNX within a few days. But the same analysis pointed to a major bottom in the ^TNX in October... so much for that method. It appears that so many technical indicators, even those found "scientifically", work a few times but then simply stop working. Nonetheless, as you know, a system only needs to produce results slightly better than chance to be very profitable in the long run. Although I am open to non-rigorous methods, I would like to see some analysis of the accuracy of the signals generated by such methods. Do you know if any accuracy analysis has been done on Bradley, time-segment analysis, or E-waves?