Keep It Simple And Complex, Stupid

Flattr this!

My last post sup­port­ing the use of non­lin­ear mod­els (“You Do Need A Weath­er­man”) gen­er­at­ed some thought­ful respons­es, main­ly along the lines of this post by Ari Andri­copou­los enti­tled “A View on the Eco­nom­ic Mod­el Debate from a Non-econ­o­mist (but some­one who builds mod­els for a liv­ing)”. The basic argu­ment is that a full non­lin­ear mod­el of any sig­nif­i­cant eco­nom­ic process would be too com­pli­cat­ed, and that it was bet­ter there­fore to stick with tractable lin­ear mod­els, while keep­ing in mind that the real world is non­lin­ear:

I build mod­els with data for a liv­ing, and I am acute­ly aware of the prob­lems with using non-lin­ear mod­els to make any sort of accu­rate pre­dic­tions – even with huge vol­umes of data to cal­i­brate it with.

It is not that the sys­tems are lin­ear. They are huge­ly com­plex. My prob­lem is that they are too com­plex to mod­el even with non-lin­ear mod­els. My belief is that lin­ear mod­els do have to be used but with a full under­stand­ing of the non-lin­ear­i­ty of real life. Also, the whole build­ing of macro-mod­els from first prin­ci­ples, based on ‘ratio­nal’ agents, is a com­plete joke of a way to design a mod­el that is sup­posed to be used in the real world.

While these points have some valid­i­ty (espe­cial­ly Ari’s jibe at “ratio­nal agent” mod­els), this crit­i­cism approach­es com­plex sys­tems from the wrong end—the “com­pli­cat­ed” as opposed to “com­plex” end. A core les­son from com­plex sys­tems analy­sis (dat­ing right from its first dis­cov­ery by Poin­care back in 1899, and man­i­fest in the first sim­u­la­tion of a com­plex sys­tem by Lorenz in 1963) is that a sim­ple sys­tem can demon­strate com­plex behav­iour. And a sim­ple com­plex system—yes, I know that sounds like an oxy­moron, but bear with me—can tell you most of what you need to know about a com­pli­cat­ed com­plex sys­tem.

Click here to read the rest of this post.

Bookmark the permalink.

About Steve Keen

I am Professor of Economics and Head of Economics, History and Politics at Kingston University London, and a long time critic of conventional economic thought. As well as attacking mainstream thought in Debunking Economics, I am also developing an alternative dynamic approach to economic modelling. The key issue I am tackling here is the prospect for a debt-deflation on the back of the enormous private debts accumulated globally, and our very low rate of inflation.