Stock Market Crashes Are Predictable; Major Decline Is Coming in 2003 and 2004, Says UCLA Physicist

UCLA ^ | 14 Dec 2002 | Stuart Wolpert

[shrinkermd]

He should read Henry Hazlitt, Ludwig Von Mises and F Hayek before he predicts stock markets and earth quakes.

[Larry Lucido]

Well, I guess a physicist should be able to predict economic trends at least as accurately as an economist.

[society-by-contract]

http://www.mises.org/

[Kozak]

If he's so smart why ain't he rich?

Lets see him put some real money on the line to back his theory.......

[rohry]

Already posted here:

http://www.freerepublic.com/focus/news/807995/posts

[richardtavor]

"Economic forecasting is often not effective at predicting changes of direction, but our algorithms are very good at doing so,"

That's exactly what a Merrill Lynch guy told me once--of course I lost a bundle.

[E=MC²]

Isn't he about 2 years LATE???

[freeper12]

>>Sornette and Zhou predict the Standard & Poor's 500 (currently above 900) will begin dropping by the second quarter of 2003 and will fall to approximately 700 in the first half of 2004.

The s&p 500 fell from over 1500 down to about 770 earlier this year...about a 50% "crash" that I guess he didn't manage to predict...now he makes the bold prediction that the markets will "CRASH" from 900 down to 700, a 20% drop or so....hmmmmmm.

[A. Pole]

Well, I guess a physicist should be able to predict economic trends at least as accurately as an economist.

He should much better as he is not blinded by an ideology (of free market or socialism or else). Physicists deal with various, complex, often not related phenomenons. Earthquakes indeed might be harder to predict.

[B.Bumbleberry]

All of these efforts rely on the false assumption that the future can be predicted by examining the past, and the erroneous belief that complicated mathmatical models can bring certainty to something that is inherently uncertain--the future economic decisions of millions of people. The study of future markets is behavioural science, not mathmatical science. So yes, a physicist can predict economic trends as accurately as an economist, which is to say, neither can.

[Keith]

yes...these idiots are so good at driving by looking in the rearview mirror.

This makes me certain a "crash" like these guys are talking about are much less likely.

[rgboomers]

The chart above is calculated with what is likely a similar algorithm to the one mentioned in the article. While the projections are not accurate enough for investment purposes, the results, in my opinion, are better than what could be achieved by chance. As my own results are encouraging, and there are persons much smarter than myself working on this problem, I agree that a revolution in stock price prediction is likely about to occur.

[Tauzero]

"He should read Henry Hazlitt, Ludwig Von Mises and F Hayek"

How do you know he hasn't?

[Tauzero]

"All of these efforts rely on the false assumption that the future can be predicted by examining the past"

Actually, this particular effort is based on three true ideas: (1) in finance there is nothing new under the sun, (2) most investors are impulsive and arational, and (3) human nature has not changed in thousands of years.

"the erroneous belief that complicated mathmatical models can bring certainty to something that is inherently uncertain--the future economic decisions of millions of people"

False charge. He doesn't claim certainty. No honest person examining the problem ever would.

"The study of future markets is behavioural science, not mathmatical science."

Aren't many behaviors, including investing behavior, quantifiable?

[Tauzero]

"If he's so smart why ain't he rich?"

The self-comforting answer to which academics subscribe is the efficient market hypothesis.

Successful trading is not primarily about intelligence.

[B.Bumbleberry]

I guess I don't see your point. The problem is with the idea that you can forecast economic results. How does the idea that people are irrational provide support. I thought that was part of the point I was making: that it is predictable that people will be unpredictable.

[Tauzero]

Two things: 1) I didn't say they were irrational, but arational. 2) Neither adjective is synonymous with unpredictable.Humans are predictably arational, or rather exhibit their arationality in predictable ways.

Basically all Sornette has done is an exercise in pattern recognition. It's almost not really forecasting at all, since he doesn't predict when a pattern will emerge.

"The problem is with the idea that you can forecast economic results."

What is the problem? I'm willing to go out on a limb and say with confidence that manufacturing activity in this country will not grind to a complete halt tomorrow. I could be wrong though. We could be hit by an asteroid or something arriving from sunward.

It's not at all a question of whether you can forecast economic results, but rather how far into the future and with what confidence interval.

[BlackJack]

When a system that works is found it is discounted and will no
longer work. Isn't that the way it works?

[rgboomers]

" When a system that works is found it is discounted and will no onger work. Isn't that the way it works? "

Yes, this is pretty much how it works. This is due to the interactive nature of markets.

When the observer is also a participant, the interactive effects (unpredictable portion) becomes large. This is what happens in the stock market all the time.

So, why is what this scientist is doing possibly different? Well, it may not be! (I am reading his book, BTW). However, there is a chance that he is getting to the actual core reason of why a certain stock market behavior (CRASHES) occurs. More importantly, it may be possible to objectively calculate when they are more likely. Granted, the previously mentioned interactive effect makes the prediction fairly wide. For example, one could never say there will be a crash tomorrow. More likely would be something like a crash is due in the next year or two. Most people would continue to find this worthless information. So, in a certain sense, the hub-bub about all this only amounts to anything if you are one of us nerdy people who study this sort of thing.

My own reasoning is that stock crashes can't occur unless stock prices get too high based on companies' potential future profits. Is that what happened in 1929, 1987, and even more recently? Yep, it is.

Basically, it comes down to this. The research in question indicates the stock market might be slightly predictable where it has always been thought to be totally unpredictable. Except to people who like to split hairs, this does not amount to much, yet.

Sincerely,
R.G. Boomers