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.

Technology Stocks : Dell Technologies Inc. -- Ignore unavailable to you. Want to Upgrade?

To: McNabb Brothers who wrote (22245)	11/20/1997 2:14:00 PM
From: Chuzzlewit	Respond to of 176387

Alright, Hank, here's a very adult post. TA is an excercise in self-delusion. It is based on the false assumption (which has been repeatedly disproven statistically) that a time series of stock prices and volumes have memory. There has never been a statistically significant demonstration that T/A works. If you are interested I could cite any number of references on this subject. Many T/A'ers rely on variants of Monte Carlo methods to prove their point, but that is erroneous reasoning. Here are a few more problems with T/A. It uses crossing of moving averages to signal reversals. There are two problems. First, the method is really a surrogate for first and second differentials. If you recall calculus, you use first differential to find inflection points, and second differentials to find maxima and minima. But to use that approach you need to have the equation for the function. But statistical analysis tells you that the function is a random walk, and therefore future movements are unpredictable. That's why you guys keep mucking around with changing the MA parameters. You are attempting to find the proper MA's ad hoc, and then apply them to future situations. The problem is that it doesn't work. Just as you categorized putting all of your money into one stock as stupid (a position with which I wholeheartedly concur), I would similarly categorize using TA for putting money at risk. Lest you think I have a closed mind on the subject, I am willing to be persuaded. Show me one statistically valid study published in a reputable journal subject to peer review that shows that T/A generates statistically significant predictive information and I will publicly eat crow in any forum of your choosing. Regards, Paul

To: McNabb Brothers who wrote (22245)	11/20/1997 2:29:00 PM
From: Chuzzlewit	Read Replies (5) \| Respond to of 176387

Alright, Hank, here's a very adult post. TA is an excercise in self-delusion. It is based on the false assumption (which has been repeatedly disproven statistically) that time series of stock prices and volumes have memory. There has never been a statistically significant demonstration that T/A works. If you are interested I could cite any number of references on this subject. Many T/A'ers rely on variants of Monte Carlo methods to prove their point, but that is erroneous reasoning. Here are a few more problems with T/A. It uses crossing of moving averages to signal reversals. There are two problems. First, the method is really a surrogate for first and second differentials. If you recall calculus, you use first differential to find inflection points, and second differentials to find maxima and minima. But to use that approach you need to have the equation for the function. But statistical analysis tells you that the function is a random walk, and therefore future movements are unpredictable. That's why you guys keep mucking around with changing the MA parameters. You are attempting to find the proper MA's ad hoc, and then apply them to future situations. The problem is that it doesn't work. Just as you categorized putting all of your money into one stock as stupid (a position with which I wholeheartedly concur), I would similarly categorize using TA for putting money at risk. Lest you think I have a closed mind on the subject, I am willing to be persuaded. Show me one statistically valid study published in a reputable journal subject to peer review that shows that T/A generates statistically significant predictive information and I will publicly eat crow in any forum of your choosing. Regards, Paul