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: Chris who wrote (44333)	5/22/1998 11:04:00 AM
From: Ashley Campbell	Respond to of 176387

I agree! I'm hoping we hit bottom around 83. We'll see. I think if the NAZ turns around, we'll head back up with it. Wish I was quick enough to play this volatility though! -ac

To: Chris who wrote (44333)	5/22/1998 1:37:00 PM
From: Chuzzlewit	Read Replies (5) \| Respond to of 176387

Chris, TA is really a bunch of hooey! There is no, I repeat NO statistical evidence to support it. ARIMA modeling clearly indicates that stock prices and and volumes have no predictive power. In case you are unfamiliar with ARIMA modeling, it is a very sophisticated modeling technique designed to generate ad hoc time-series models. I have repeatedly challenged TAers to cite one reputable statistical study that demonstrates the efficacy of the approach. Let me make some specific points. Trend analysis depends on where the trend is assumed to have begun. So if you go back to Dell's last earnings report you will find that the stock was around $69. So, if we're trading today at around $85 that represents a 23% increase (or 130% annualized). Now, if you choose to look at the high a couple of weeks ago, that represents a decrease of 13.3%, or an annualized decrease of 97.5%. The point I'm making is extrapolating trends is nonsense. So, TA resorts to utilizing the argot of statistics to justify a statistically unsound system. We have the use of moving averages as extrapolative tools, and then to approximate inflection points (remember your differential calculus?) they use fast and slow MA's. Unfortunately, since the underlying function can't be discerned, how can you predict an inflection point? Ah, details, details ... But wait, it gets better. In statistics, we use the concept of standard deviation, where an estimate is given for a normally distributed population, plus or minus 1 std. deviation includes around 67% of the observations. So what does the TAer do? He comes up with a Bollinger band as a surrogate. BUT THERE IS NO MATHEMATICAL JUSIFICATION FOR THIS !!!!! So, neither from a mathematical/statistical point of view nor from an empirical (statistically demonstrable) point of view does this approach make any sense. I do have an open mind. If you can cite any evidence that this stuff works, I will be happy to read it. I am not interested in anecdotes -- only statistically verifiable studies that have not engaged in data mining! In the mean time, I would just as soon consult the psychic hot-line or my local dowser, or engage in haruspex as consultant TA for investment opinions. But that's just my opinion! TTFN, CTC