The world of ABMs…

No, dear Reader, we are not referring to the MAD world of Dr. Strangeglove and anti-ballistic missiles (ABMs), but to the origins and concept of Agent-Based Models (also ABMs). The former are irredeemably destructive; the latter the basis for much of modern forecasting and risk assessment.

As regular readers will know, we pay attention to the work of Andrew Haldane, Chief Economist of the Bank of England- a man who is something of an iconoclast, as well as a skillful disseminator of ideas. So, we read with interest a speech he gave recently entitled “The dappled world”. The title is a reference to the work of a philosopher of science, Professor Nancy Cartwright, in which she made the point that the so-called “natural sciences” are in reality the product of a “patchwork of theory and evidence” (rather than some wonderful idealized Platonic form.)

Now, one would think that this would be considered “a statement of the obvious”. The world is a messy place, and even some “iron laws” of, say, physics are not quite a rigid as one might think- compare the Newtonian world with that of quantum mechanics. As Haldane points out, CERN’s huge and complex Large Hadron Collider is “in essence, a massive Monte Carlo machine”. If we throw spaghetti against the wall in sufficient quantities and frequently enough, the probabilities are that at some point we will think we see a pattern!

And you might be surprised where and how ideas from elsewhere can infiltrate economic and financial theory and practice. The (in)famous Black-Scholes Formula for options pricing had at least part of its origins in the heat transfer equation in physics: Haldane describes it as a “genetic mutation”.

However, the key question in all this is: “What works?” Economic models and theories are all very well, but they lead remarkably often to failure in terms of outcomes, with unpleasant consequences for those who are impacted. Of course, having no model or frame of reference can be as bad as having one that is flawed. One has to start somewhere; otherwise we would run the risk of not being able to decide upon anything by suffering from intellectual paralysis.

And that brings us back to ABMs. Somewhat ironically, these grew out of the work of Enrico Fermi at the Los Alamos National Laboratory during the Manhattan Project, when he was trying to calculate how to model the chain reactions of fissioning atoms when helping to design the first atomic bomb. In time, this work led to less destructive applications across an ever-increasing array of industries including finance, where the uses of random numbers and of “agents” who are both heterogeneous and interactive are now fundamental to the models built to predict outcomes and scenarios. In another delicious irony for the realm of (re)insurance, Monte Carlo simulations, of course, take their name from the eponymous Casino.

So, what has all this to do with Awbury? While the Team aims to operate in a cohesive and effective manner, and has, in the 5 years in which we have been in operation, developed a recognizable “culture”, it is in reality a group of individuals whose views can certainly be considered heterodox, because we are very well aware of the dangers of “groupthink”. Yet, at the same time, we know that to make sense of and structure the opportunities we have it is essential to build and use effective models if we are to make judgements and decisions that are realistic in risk/reward terms. Yes, the world is messy and human beings can do remarkably stupid things judged objectively, but that is the reality; there are limits and constraints; experience matters; and, as long as one avoids being captivated by the beauty of the models one has built, one has a reasonable chance of making an executable and profitable decision- something on which our partners rely.

In truth, we all live in a world of ABMs- of both varieties.

The Awbury Team


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.