Swans, market change and uncertainty

Greetings (almost festive according to John Lewis) once again to our blog. Over the past few weeks I’ve been inspired in various senses by philosophy and physics. So before that sounds too pretentious let me explain.

Firstly – swans…. Many of you may have read Talebs book ‘the Black Swan’ which in very brief summary is an account of how very unlikely events do not form a part of the equations, or mental models that we use to represent the world, and hence when they do occur can have very large impacts on the way the world works. I happened to be reading a philosophical work on holiday recently and chanced across a description of induction by Bertrand Russel, in which he described the inductive process very well. When you or I see a swan it is white (almost without precedent – I have seen a black swan once) hence when you describe a swan to someone else, its most likely that you would tell them that swans were white. So a black swan would come as a real surprise. This is an example of inductive thinking, where your expectations are set by what you recently experienced. Where things are expected to happen in the same way that they previously happened. Induction in this sense is new term to me, as induction means creating electricity with coils of wire and a magnet … Inductive thinking is an extremely powerful process, which has allowed us to develop an enormous number of modelling techniques, and the fields of mathematics.

Secondly – new thinking about gravity…. This article from the Guardian provide me with real hope that we can stop flogging the dead horse of string theory and find something new. I was really interested in the comparison of the general theory of relativity here to our quantum theories. The first is a theory of unification, where every body effects every other. This implies the interconnectedness of all things (like in Dirk Gentlys holistic detective agency). My pencil has an effect on your piece of paper when I write on mine. This impact diminished with distance and mass (so don’t worry too much). Quantum theory on the other hand is a theory which only really works on a microscopic level…. This is the limit of my understanding, so I won’t continue at risk of embarrassing myself, but if you’d like to know more please read the excellent article http://www.theguardian.com/news/2015/nov/04/relativity-quantum-mechanics-universe-physicists.

What I observe from these processes, and from the use of regression in econometrics, or trading, is that when we form statistical models (which are models by induction in pure form), we do so ignoring the rest of the world which takes place around our microcosm. Our estimates are made in a specific sense about the data we choose to observe, in relation solely to one another. What if our error terms included both the known and unknown? What if we were able to estimate the chances of a model being correct if we could expand our sample to cover the world, or the Universe. Would this effect how we viewed our models, or our results? I’m sure that someone very clever has already come up with a theory which deals with this, but I haven’t found it yet… So please let me know if there is one.

OK, so lets hope you aren’t bored, mystified or both so far. How does this effect ChargeSync? And why am I ruining your breakfast? We speak now of changing worlds, new paradigms, a fresh universe. Old equations do not apply, and the unknown universe (and hence error) is much bigger. It takes people, and minds time to adjust to a new world and new philosophy. That adjustment is taking place today as people imagine a world without fossil fuel generation (or with it only to a limited extent).  So used are we to the old models, and so much comfort do we have in our old ways of thinking, that the potential is to provide one more recursive step in the inductive loop. But what once was a white swan will soon emerge not as a black swan, but as a giant orange badger with a hat on (something a long way removed from a swan is what I’m trying to say). Try putting that in your model.

How do we deal with that? Use your imagination, trust your instincts, discard history – but be aware of it. And induct at your peril. Good luck!