An economics, investment, trading and policy blog with a focus on Modern Monetary Theory (MMT). We seek the truth, avoid the mainstream and are virulently anti-neoliberalism.
Pages
▼
Pages
▼
Wednesday, April 23, 2014
Mark Buchanan — Economic orthodoxy is going down … even if most economists don’t yet know it
Mark Buchanan makes the very debatable point that modelling with supercomputers will improve economics. The big and blindingly obvious problem there is that there is a huge lack of agreement as to what to feed into the computer. E.g. do we give it an MMT view of economics or a market monetarist view or what?
I agree with Mark that DSGE models are a dead end. I admit that I am not familiar with agent-based modelling, but I would guess that they have the same problem that overlapping generations models have - small tweaks to the "micro" behaviour mean large changes to the "macro" behaviour of the model.
This is a great technology for a video game, or for teaching models. You are telling a story about a fictional world.
But it will be very hard to impossible to fit to real-world data. But that is what the mainstream wants to do. To do that, you need to stick with an aggregated model, like a Stock-Flow Consistent model.
Mark Buchanan makes the very debatable point that modelling with supercomputers will improve economics. The big and blindingly obvious problem there is that there is a huge lack of agreement as to what to feed into the computer. E.g. do we give it an MMT view of economics or a market monetarist view or what?
ReplyDeleteI agree with Mark that DSGE models are a dead end. I admit that I am not familiar with agent-based modelling, but I would guess that they have the same problem that overlapping generations models have - small tweaks to the "micro" behaviour mean large changes to the "macro" behaviour of the model.
ReplyDeleteThis is a great technology for a video game, or for teaching models. You are telling a story about a fictional world.
But it will be very hard to impossible to fit to real-world data. But that is what the mainstream wants to do. To do that, you need to stick with an aggregated model, like a Stock-Flow Consistent model.