Physicists Successfully Predict Stock Exchange Plunge

[churchill]

WITH 20/20 hindsight, financial crashes seem inevitable, yet we never see them coming. Now a team of physicists and financiers have bucked the trend by successfully predicting a steep fall in the Shanghai Stock Exchange.

Their model, which employs concepts from the physics of complex atomic systems, was developed by Didier Sornette of the Financial Crisis Observatory in Zurich, Switzerland, and Wei-Xing Zhou of the East China University of Science and Technology in Shanghai. The idea is that if a plot of the logarithm of the market's value over time deviates upwards from a straight line, it's a clear warning that people are investing simply because the market is rising rather than paying heed to the intrinsic worth of companies. By projecting the trend, the team can predict when growth will become unsustainable and the market will crash.

Sornette, Zhou and colleagues applied their model to the Shanghai Composite Index, which tracks the combined worth of all companies listed on the Shanghai Stock Exchange, the world's second largest. Early this year, the index gained 50 per cent in just four months. In July, the team predicted that the index would start to fall sharply by 10 August (www.arxiv.org/abs/0907.1827). The index duly began to slide on 4 August, falling almost 20 per cent in the subsequent two weeks.

Anyone hoping to exploit the model for profit should think twice. "If enough investors take action based on our predictions, the evolution of prices will probably be affected," says Zhou.

http://www.newscientist.com/article/mg2032...nge-plunge.html

[lannarebirth]

WITH 20/20 hindsight, financial crashes seem inevitable, yet we never see them coming. Now a team of physicists and financiers have bucked the trend by successfully predicting a steep fall in the Shanghai Stock Exchange.

Their model, which employs concepts from the physics of complex atomic systems, was developed by Didier Sornette of the Financial Crisis Observatory in Zurich, Switzerland, and Wei-Xing Zhou of the East China University of Science and Technology in Shanghai. The idea is that if a plot of the logarithm of the market's value over time deviates upwards from a straight line, it's a clear warning that people are investing simply because the market is rising rather than paying heed to the intrinsic worth of companies. By projecting the trend, the team can predict when growth will become unsustainable and the market will crash.

Sornette, Zhou and colleagues applied their model to the Shanghai Composite Index, which tracks the combined worth of all companies listed on the Shanghai Stock Exchange, the world's second largest. Early this year, the index gained 50 per cent in just four months. In July, the team predicted that the index would start to fall sharply by 10 August (www.arxiv.org/abs/0907.1827). The index duly began to slide on 4 August, falling almost 20 per cent in the subsequent two weeks.

Anyone hoping to exploit the model for profit should think twice. "If enough investors take action based on our predictions, the evolution of prices will probably be affected," says Zhou.

http://www.newscientist.com/article/mg2032...nge-plunge.html

I'm always looking for the physics of the market as well. I'm thinking phugoidal descent.

Anyhow, those researchers could have checked to see if there was ever a a parabolic move in any market in any era that didn't retrace 100%.

[LaoPo]

If you predict a sharp fall of the Shanghai Index starting on August 10 and it started 6 days earlier on August 4.....

..........are you then successful ?

LaoPo

[lannarebirth]

If you predict a sharp fall of the Shanghai Index starting on August 10 and it started 6 days earlier on August 4.....

..........are you then successful ?

LaoPo

You're only successful if your bets have been layed, but timewise from an academic perspective that's certainly close enough.

Edited August 28, 2009 by lannarebirth

[churchill]

But they are saying that nobody can know about it otherwise it will not work -Similar to the Goldman Sachs program http://www.bloomberg.com/apps/news?pid=206...id=ajIMch.ErnD4

[badge]

Sornettes work been about for seemingly ages. this is interesting on behavioural science

http://audiovideo.economist.com/

[Junglejumbo]

This is very risky. Elliot's wave theory is probably more reliable. It's based on natural concept of number patterns that occurs in tree leafs, sea waves, the human body.

I am just thinking that if a price deviates from a straight line (presumably an upward straight line) then how do you know it's not going to rise even further....you won't know exactly when to sell. Also even though at first it looks like it's deviated much, over time you could see another higher line that can be drawn under the higher prices--this conforms with Elliot wave theory of five waves up and two down before either another cycle up or collapse. What if you already sold? You may have to then wait years if ever for the that line or shall we say "support" to be broken before it falls down again to the line originally drawn. I think a market like the Shanghai composite is a very volatile market and August is a volatile month due to summer vocations in the West, and the start of portfolio adjustments before which usually continue on through September.

[AlexLah]

Finally some kind of confirmation that it is sort of possible (it seems) to calculate a significant downward movement. And again not to brag about it but I have called the October crash (first two weeks of Oct I mentioned) way before, and the Feb 9 crash (way before) even to the exact date (some time before), and another small down (4%) move in May I believe it was.

Again I say, I am not someone that trades, or invest in the stock market, I just wanted to know if I could discover some patterns during my 11 months being unemployed and bombard this board with silly pictures in several threads and posting some cryptic and insane messages, ha ha ha!

On the other hand sometimes I posted something, numbers of a pattern that I found like:

QUOTE (AlexLah @ 2008-10-08 16:56:10) *

I have told you in the past and I will tell you in the future.

The first and second week, my words have been marked. (I pointed at October)

Follow the money and understand the 7th principle. ( you can Google this to understand)

13 is the number and the next is 7. (Let me explain)

(The colored comments I have added)

-------------------------------------------------------------------------------------------------------------------------

13+7=20 we have added so now we have to deduct, 2-0=2

December is the 12th month, so minus 2 is month 10 which is October.

As we have deducted we have to add again to restore imbalance, so 1+0= 1, Ying Yang you know.

So I concluded it would be the 1st week of October and possibly including the second for a downfall as 2+0=2 referring to the first line.

(Go from months to weeks to days)

(I now understand it was clearly pointing at the 1st of October but I was not sure at that time, that came later)

---------------------------------------------------------------------------------------------------------------------------------------------------------------

13 is the number and the next is 7.[/i] (Let me explain)

3-1=2, 2+7=9, so pretty easy: 02-09 would be the next downturn. First I was a bit confused as I wanted to use the EU style of date like 02-09 would be 2 September but then I realized it should be US style as the Dow is from US and US always writes down the month first.

----------------------------------------------------------------------------------------------------------------------------------------------------------------

You see I started with adding and the next one was about deducting first, it is all about balancing and following the rules.

----------------------------------------------------------------------------------------------------------------------------------------------------------------

I am sure some of you will call me a lunatic and crazy but that is OK, I am used to it here. The question for you is: Where or how did I came up with these numbers and calculations?

Take care all,

[badge]

Finally some kind of confirmation that it is sort of possible (it seems) to calculate a significant downward movement. And again not to brag about it but I have called the October crash (first two weeks of Oct I mentioned) way before

In all honesty I 'scan' through most of your posts Alex, as in the past theres been lots of conspiracy and manipulation posturing, and thats of little interest to me. Yet I too believe that financial markets may see some turbulence soon, and that the US' SPX Index could find itself around 850 by the end of October. Or indeed, if the other way, perhaps 1090

Not too far from what you suggest.

I have to say im not convinced enough by my work to 'bet' on it solely, but if price action looks interesting around there i'll certainly short it.

Lets see

[emsfeld]

WITH 20/20 hindsight, financial crashes seem inevitable, yet we never see them coming. Now a team of physicists and financiers have bucked the trend by successfully predicting a steep fall in the Shanghai Stock Exchange.

Their model, which employs concepts from the physics of complex atomic systems, was developed by Didier Sornette of the Financial Crisis Observatory in Zurich, Switzerland, and Wei-Xing Zhou of the East China University of Science and Technology in Shanghai. The idea is that if a plot of the logarithm of the market's value over time deviates upwards from a straight line, it's a clear warning that people are investing simply because the market is rising rather than paying heed to the intrinsic worth of companies. By projecting the trend, the team can predict when growth will become unsustainable and the market will crash.

Sornette, Zhou and colleagues applied their model to the Shanghai Composite Index, which tracks the combined worth of all companies listed on the Shanghai Stock Exchange, the world's second largest. Early this year, the index gained 50 per cent in just four months. In July, the team predicted that the index would start to fall sharply by 10 August (www.arxiv.org/abs/0907.1827). The index duly began to slide on 4 August, falling almost 20 per cent in the subsequent two weeks.

Anyone hoping to exploit the model for profit should think twice. "If enough investors take action based on our predictions, the evolution of prices will probably be affected," says Zhou.

http://www.newscientist.com/article/mg2032...nge-plunge.html

I'm always looking for the physics of the market as well. I'm thinking phugoidal descent.

Anyhow, those researchers could have checked to see if there was ever a a parabolic move in any market in any era that didn't retrace 100%.

They havent given any details so one can just guess. Their description sounds like those guys simply logtransformed Shanghai Composite Index prices, regression fitted a trend line, removed that trend line from log prices and then obtained the logresiduals. Sounds like the logresiduals display some mean-reverting behaviour so that they were able to empirically identify a few reversal thresholds (e.g. TAR type models). Though I do believe market prices mean-revert to an exponential trend over a very longterm time frame, however, I am not sure to what extent this information is useful to someone in it for investment horizons of six months or less. Maybe the guys were just lucky or maybe the Shanghai Composite Index indeed has this property over relatively short cycle periods. Interesting nevertheless!