Have you ever been thinking that there is something fundamentally wrong with the way economists view the world? One reason is that Economists apply linear and analytical models which simply don’t take into account e g non-linearity, which is exactly what we see around us today. During certain periods the world seems to behave according to linear models and things seems almost predictable, but between those stable periods non-linear or even chaotic phenomena is dominating and linearly created models becomes totally irrelevant.
This video is an interview with Doyne Farmer, a professor doing research complex systems at the Santa Fe institute. In this video he explains his project of making a bottom-up agent-based simulation of the economy.
The really interesting value with agent-based simulations is that they don’t reduce problems into a linear models. On the other hand they require calibration to reality as well as a lot of computer power to simulate the actions of millions of individuals.
What we also must know is that they are not very good at predicting future outcomes. They rather help is to learn about how certain patterns appears and how these patterns have certain emergent results. It is a tool for learning rather than predicting.
One way to understand this is to see the computer as a “macroscope” which helps us zoom in and out from individual behavior to the emergent patterns in order to understand under which circumstances e g collapses and other phenomena happens.
This emerging understanding and research of non-linearity, complexity and chaos have been interesting for scientist from many field for a couple of decades now, almost nothing of it have seems to have been applied to e g econometry. It is really urgent that this happens now and professor Doyne Farmer seems to be the right person to take the first steps towards a new complexity based theory of economics.
via Doyne Farmer - Macroeconomics From the Bottom Up | Institute for New Economic Thinking, futuramb
(via emergentfutures)
