Didier Sornette: :


Source: Ted

Category: Cycles, Video

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

20 Responses to “How We Can Predict the Next Financial Crisis”

  1. digistar says:

    I wish Didier and his team would make their ‘correction’ predictions public, before the corrections occur, so we could take corrective action – individually – if not also as a species.

    Realistically, I think it unlikely that, as a species, we can take corrective action. The 1% of us who benefit from the status quo will prevent it.

    • emailcraigs says:

      Exactly, the current “system” did not come into being by accident. Money created it in order to create more money.

  2. b.remson says:

    I’m guessing that this will not be an open source model?

  3. gman says:

    The premise may be wrong here…the booms and busts could be a feature not a bug to the people in control of the system…not to be controlled but exacerbated.

  4. mpavan says:

    Interesting: yet another dynamical analysis suggesting that global society will reach a ‘transition stage’ or ‘turning point’ or ‘start of crash’ around 2050. The first was The Club of Rome “limits to growth’ in 1972 (confirmed in 2008 http://www.csiro.au/files/files/plje.pdf) and I’ve seen others pop up.

    Dynamical analyses are the ONLY way to hope to understand time-dependent complex systems. That is why the vast majority of economists (save Steve Keen and others using DA) are completely out to lunch for anything but very short-term look-out-the-window type analyses of what is going on.

    Sadly for my kids, Monsieur Sornette and others will be ignored.

  5. swag says:

    Oh, god, this guy again? Does anybody remember how well his stock market predictions (based on his rocket science) circa 2004 turned out? Hint: not very well.

    What’s the name of that behavioral foible, again, where one mistake’s one’s expertise in one area for expertise in an unrelated area?

  6. pielou says:

    Who will be in charge to decide if the dragon has to be killed or not? what future do we change by not letting it run its course?
    We just have to invent time travel now.

  7. Angryman1 says:

    Look at leverage in any such market. Now, there are “bubbles” like in the 90′s which was due to capital freeing from the computer revolution. Digitalization caused interest rates in private markets to fall very low which lead toward a lot of price reduction on purchases.

    The 00′s were the opposite. The revolution ended and real interest rates soared in private local markets while leverage rocketed in 99-01 in the primary banking system as money poured in from Asia while local markets overheated. This capital excess should have been dealt with then. Instead, they allowed it to fester until the primary financial mechanisms were flush with cash.

    The coming boom looks like the 90′s except the computer revolution productivity has been replaced by Euroflight. Huge capital flight from Europe due to the Euro crisis is driving real interest rates rock bottom in local markets, sucking up capacity, boosting production in the US economy, thus keeping the world stable. Sadly, if you are in the European South, you may not be so happy………….

  8. lblatt says:

    The granddaddy of this sort of analysis was the late Benoit Mandelbrot — better known to the general public, if at all, as the inventor of “fractals,” and the first person to use computers to help visualize the strange behavior of these often very simple mathematical objects. Mandelbrot published many research papers on the modeling of financial markets, along with a lot of other human-driven time series, beginning in the late 1950′s and throughout the ’60′s and ’70′s. He summed up his many decades of studies of the financial world in a book called “The (Mis)Behavior of Markets: A Fractal View of Risk, Ruin, and Reward,” published in 2004. Didier Sornette’s related work in this area was published under the title of “Why Stock Markets Crash: Critical Events in Complex Financial Systems,” also in 2004. Sornette’s is easier reading, and more directly of interest, I would guess, for most investors who think at all along “technical” lines.

  9. Willy2 says:

    “How to predict the next financial crisis”

    Simple. Look at what happened in the past. It provides a guide to what is going to happen in the (near) future.

  10. Bridget says:

    Ha! I came of age in the early 80′s, a time during which, judging by the graphs, we had it even worse than today’s Millenials.

    That being the case, I can attest that ignorance is bliss. We had no idea how lousy we had it, and were much the happier for it.

  11. CitizenWhy says:

    Two “understandings”:

    1. Labor share of wealth will keep going down. The gradual grinding of the majority into unemployment or underemployment (part time, no benefits), lifelong substantial debt, loss of assets, or lower real income. The labor sector of the economy will continue shrinking in wealth, income, and political influence. The shrinking, on the whole, will be gradual enough to tamp down any rebellion.

    2. Recessions/depressions enable the 1% to buy land and other assets at dirt cheap prices. The 1% like crashers. If you come to own the entire world it does not matter what the world’s GDP is.

  12. jacobsk says:

    ECRI has some model and it worked for a while and now they don’t even know what’s wrong with their model. This guy has some secret model it may work for a while and then it won’t. Then there’s the Hindenberg Omen model, which has been leaking ever since it first occurred.
    The only model that comes close to perfection is the Hurricane model, which combines a bunch of models and gives the forecast cone.

  13. Mattw says:

    Predicting the timing of collapses is very difficult. One may have to settle for getting within a few years.

    Below is a summary of how collapses work. The links have been stripped out.

    Collapse Framework for Societies

    What follows is a framework for viewing collapses in society – economic collapse or war/revolution.

    1. Societies follow a positive feedback loop process. Each new day is heavily dependent on the past. This is similar to forests and sandpiles. The process never stops and collapses are impossible to prevent. One may only transform the size and timing of collapses.
    2. Positive feedback loop processes are subject to self-organizing criticality. They will automatically move toward a pre-collapse state, then just collapse.
    3. Collapses follow a power-law distribution. Outliers exist.
    4. All collapses are the same. There is no difference (other than size) between a small collapse and a big collapse. Big collapses require longer to form and happen less often.
    5. Collapse transformation: Collapse suppression will delay a collapse and make the resulting collapse bigger. Suppress small collapses and you will get bigger collapses. Suppress bigger collapses and you will get the mother of all collapses. Suppress that and you will sit on the edge of a cliff forever waiting to fall or be pushed over the edge. Think Japan.
    6. Collapses are caused by the build-up of bad ideas, bad decisions and corruption. These things can spread to all corners of society.
    7. War or revolution is just a collapse like an economic collapse. Only the form is different.
    8. Collapse suppression leaves the original problems (bad ideas, bad decisions and corruption) in place giving them the ability to continue growing.
    9. A collapse in one area could mean problems in other areas as well. For example, 9/11 could mean more than a terrorism problem. It could represent a sign of spreading problems into all corners of society.
    10. The longer the time of stability means the longer (and bigger) the problems can grow. Time of relative stability is the most important criteria in determining when a large collapse is possible. History helps us determine which time frames are important.
    11. When a system has reached a point where a small event can have a large impact then it is at a pre-collapse state or tipping point. Causality is not linear.
    12. Big collapses (the outliers) may represent phase transitions where everything you know changes.
    13. Black swans are outliers in a normal distribution which cause a phase transition. Unpredictable.
    14. Dragon-kings are outliers in a power-law distribution which cause a phase transition. Predictable.
    15. Examples of systems with a power-law distribution (outliers allowed): Forest fires, sandpile collapses, snow pile avalanches, earth quakes, financial market collapses, wealth, city size, serial killers, riots, attacks within war and wars.
    16. In financial mathematics, the use of the normal distribution is forbidden. It assumes behavior is independent. In a crisis or collapse, market behavior is not independent as people start herding. Naturally this means in reality all financial mathematics uses the normal distribution. Were you wondering why these models blow up?

  14. Singmaster says:

    Digistar,
    Sometimes they do post the info in advance. 21 May they posted SPY was about to crash.
    http://www.er.ethz.ch/fco/index
    He sells his system to hedge fund managers so fund the institute. If you have $25k available, you can buy it too. He likens it to flying a fighter jet. It takes time to master.
    There is a Chinese hedge fund using it. I don’t know who else.
    He is working on a platform that will be available for small investors but he has been working on it for over a year.

  15. snowflake says:

    Let me get this straight, Barry. You spend lots of time telling us there are no gurus, which is the truth, but then the umpteenth academic comes along with a “matematical model based on the hard sciences”, and somehow he’s an exception.

    Please explain.

  16. GP says:

    Re. your link to Sornette’s TED talk today:

    It’s a reminder of how far academics are from the trenches :-)
    Sornette has his own defintion of a “bubble”, which I think is hyperbolic (pun?).

    But let’s say his “bubbles” equates to “ripe for a short term, significant drop”,
    because that’s how he actually tests his hypotheses.
    His reported results are on p.15 of the attached “fbe…” pdf.
    Briefly, they are poor.
    Furthermore, the “second strategy” for trading (detailed on p.14) is laughably hypothetical and post-hoc.
    He’s says he’s not cherry-picking, but proceeds to egregiously do so!

    More recently, see p.7 of the second attached pdf from Sornette.
    This chart shows a poor forecasting record:
    1) Missed the runups- downward spikes indicative of “antibubble” are absent.
    For a practitioner, this is unacceptable.
    2) Restricting our analysis only to his guidance to short market exposure
    during large positive spikes of endogeneous risk [chart], we find:
    -a miss in May 2010,
    -six months early in Feb 2011. Half credit only.
    -a hit in April 2012,
    -a major miss (so far) in Jan 2013. Five months early and counting.

    Suppose you or I had gone net flat (or as stated by the model, short) based on the Jan 2013 spike in “qualified bootstraps”;
    I could be out of business… and subject to ‘bad fiduciary’ lawsuits.

    Didier’s been doing this for many years. Hedgies would happily pay him a nice
    consulting fee if he had top-notch real-time forecasting skill.

    Greg