Fuzzy Logic and Predictions for an Economy

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #20484

    I have always been fascinated by Fuzzy Logic and it’s ability to work with imcomplete and noisy data. It was all the rage in electronic control systems some time back, but the “believes” in the more standard PID control systems beat back the Fuzzy folks and I don’t seem to see much being done with it.

    But as I listen to some of the lessons on the history of economic thought, it is very easy to see that the economy is complex….and quite frankly, while we might know more these days, I might even question that we know all that much of the total because there really are distinct and “different” camps and for good reason…..none of the camps seems to understand (in terms of prediction) all the particular twists and turns of the economy. (eg: what we now call the “new normal” is so called because it is a bit of a mystery what will happen next)

    Anyways…..like the weather, the economy is a complex system, and like the weather, if we imagined we actually knew all the “rules / laws” of operation, we don’t have enough “sensors” in enough places to know exactly what comes next. (and I question that we even know all the interrelationships and equations) This sounds like the incomplete / noisy data that Fuzzy Logic was built to deal with.

    Also then, since Austrian Economists love to map out and explain outcomes by looking at the result of purposeful human behavior (knowing that while there might be certain things that “tend” to be true, even that is a noisy transfer function. I would think that Austrians would be drawn to pulling together a set of Fuzzy rules, that attempts to explain (predict) future economic conditions based on at a set of monetary / human actions as inputs. (i might not be explaining this very well…..ok, perhaps the word MAYBE is an understatement)

    Do does anyone know of a group or person working on this sort of model?

    My Google search of “fuzzy logic of world economy” shows activity and discussion…..even a book or two. The work seems to span from 2002 through 2016….but I’ve not heard much spoken on this topic / idea.

    Any comments / ideas…..

    -Quentin Lewis

    #20485
    gerard.casey
    Participant

    Dear Mr Lewis,

    I’ve just come across your contribution of last November (!) and, as no one else has responded, I thought I’d make a few comments.

    Like you, I have been fascinated by the idea of a logic which would deal with relationships between classes whose boundaries are less rigid and demarcated than those with which traditional logic, in its various varieties, concerns itself. This has particular application to the Sorites argument which suggests that if you can travel from A to B on a continuum, then there is no real difference between what one can find in the continuum’s extremes.

    I think that you are probably right in thinking that the actual economy is characterised by ‘noisy data’ which makes it a prime candidate for a fuzzy logic, if ever there was one.

    However, I don’t know of any economists, still less Austrians, taking such an approach.

    Your contribution has stimulated me to go back and take a look, once again, at this material.

    I wish I could have been more helpful but your comments/questions have reawakened my interest.

    Best wishes,

    Gerard Casey

Viewing 2 posts - 1 through 2 (of 2 total)
  • You must be logged in to reply to this topic.