Getting Techie

by johnmccreery

What’s past is prologue. Serendipitously, this is a fundamental idea in the way in which Miller and Page, the authors of Complex Adaptive Systems introduce a more formal approach to modeling, based on the work of Holland, et. al., on artificial learning systems.

Step 1

We assume that the real world consists of various states, S, and a transition function F(S) that maps a given state at time t into a new state at time t+1.

The problem, they point out, is that the real world is a complicated place. The model builder starts out with at best rough ideas about what S and F(S) might be. The modeler would like to discover F(S), but

this will typically be impossible given the dimensionality of the state space and the potential complexity of F(S).

Consider the problem of predicting the weather or the stock market over a span of more than a few days. Our ideas about the relevant dimensions of the space and how critical factors might change from one moment to the next are at best approximations to what is going on. So,

Step 2

As an alternative, the modeler can reduce the size of the state space and seek a simpler transition function, substituting a model state s for the actual state S and a simplified transition function f(s) for the actual transition function F(S).

Thus, as a practical matter, the meteorologist looks at a few basic measurements, barometric pressure, temperature and humidity, for example, and bases predictions on simple assumptions about their relation to the probability of precipitation and the form it is likely to take. The stock market analyst works in a similar way, focusing on what he or she takes to be critical indicators affecting movements in the market. I put “takes to be” is in scare quotes to point to what should be evident to any reader of financial news, which tends, like all news, to be event driven, with analysts scrambling to figure out what may have been behind the event in question before the news cycle makes it only yesterday’s news.


One Comment to “Getting Techie”

  1. The stock market analysts I hung out with when I lived in the SF Bay area during the 90s were mostly complete numbskulls, but because in those days everything they touched turned to gold, they all thought they were freaking King Midas.

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

%d bloggers like this: