Skip to Content

Econophysicist Accurately Forecasts Gold Price Collapse

The first results from the Financial Bubble Experiment will have huge implications for econophysics.

There are good reasons to think that stock markets are fundamentally unpredictable. Many econophysicists believe for example, that the data from these markets bear a startling resemblance to other data from seemingly unconnected phenomena, such as the size of earthquakes, forest fires and avalanches, which defy all efforts of prediction.

Some go as far as to say that these phenomena are governed by the same fundamental laws so that if one is unpredictable, then they all are.

And yet financial markets may be different. Last year, this blog covered an extraordinary forecasts made by Didier Sornette at the Swiss Federal Institute of Technology in Zurich, who declared that the Shanghia Composite Index was a bubble market and that it would collapse within a certain specific period of time.

Much to this blog’s surprise, his prediction turned out to be uncannily correct.

Sornette says there are two parts to his forecasting method. First, he says bubbles are markets experiencing greater-then-exponential growth. That makes them straightforward to spot, something that surprisingly hasn’t been possible before.

Second, he says these bubble markets display the tell signs of the human behaviour that drives them. In particular, people tend to follow each other and this result in a kind of herding behaviour that causes prices to fluctuate in a periodic fashion.

However, the frequency of these fluctuations increases rapidly as the bubble comes closer to bursting. It’s this signal that Sornette uses in predicting a change from superexponential growth to some other regime (which may not necessarily be a collapse).
While Sornette’s success last year was remarkable it wasn’t entirely convincing as this blog pointed out at the time

“The problem with this kind of forecast is that it is difficult interpret the results. Does it really back Sornette’s hypothesis that crashes are predictable? How do we know that he doesn’t make these predictions on a regular basis and only publicise the ones that come true? Or perhaps he modifies them as the due date gets closer so that they always seem to be right (as weather forecasters do). It’s even possible that his predictions influence the markets: perhaps they trigger crashes Sornette believes he can spot.”

That’s when Sornette announced an brave way of test his forecasting method which he calls the Financial Bubble Experiment. His idea is to make a forecast but keep it secret. He posts it in encrypted form to the arXiv which time stamps it and ensures that no changes can be made.
Then, six months later, he reveals the forecast and analyses how successful it has been. Today, we can finally see the analysis of his first set of predictions made six months ago.

Back then, Sornette and his team identified four markets that seemed to be experiencing superexponential growth and the tell tale signs of an imminent bubble burst.

These were:the IBOVESPA Index of 50 brazillian stocks, a Merrill Lynch Corporate Bond Indexthe spot price of goldcotton futures
These predictions had mixed success. First let’s look at the failures. Sornette says that it now turns out that the Merill Lynch Index was in the process of collapse when Sornette made the original prediction six months ago. So that bubble burst long before Sornette said it would. And cotton futures are still climbing in a bubble market that has yet to collapse. So much for those forecasts.

However, Sornette and his team were spot on with their other predictions. Both the IBOVESPA Index and the spot price of gold changed from superexponential growth to some other kind of regime in the time frame that Sornette predicted. That’s an impressive result by anybody’s standards.

And the team says it can do better. They point out that they learnt a substantial amount during the first six months of the experiment. They have used this experience to develop a tool called a “bubble index” which they can use to determine the probability that a market that looks like a bubble actually is one.

This should help to make future forecasts even more accurate. Had this tool been available six months ago, for example, it would have clearly showed that the Merrill Lynch index had already burst, they say. If Sornette continues with this type of success it’s likely that others will want to copy his method. An interesting question is what will happen to the tell tale herding behaviour once large numbers of analysts start looking for and betting on it.

It’s tempting to imagine that this extra information would have a calming effect on otherwise volatile markets. But the real worry is that it could have exactly the opposite effect: that predictions of the imminent collapse whether accurate or not would lead to violent corrections. That will have big implications for econophysics and those who practice it.

Either way, Sornette is continuing with the experiment. He has already sealed his set of predictions for the next six months and will reveal them on 1 November. We’ll be watching. Ref: The Financial Bubble Experiment: Advanced Diagnostics and Forecasts of Bubble Terminations Volume I

Keep Reading

Most Popular

This new data poisoning tool lets artists fight back against generative AI

The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models. 

Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist

An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.

Data analytics reveal real business value

Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.

The Biggest Questions: What is death?

New neuroscience is challenging our understanding of the dying process—bringing opportunities for the living.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.