Monday, June 24, 2013

A nice piece for people who flip out when economists talk about optimizing agents

Botanists say that plants solve optimization problems too!!! And then they model it!!!

Of course, plants don't actually optimize. They have some evolved, biological heuristic they use to approximate optimization. But we model that mathematically and speculate about what they're doing because that is a useful way of trying to understand what's going on.

When people get upset about models that use optimization, it's a pretty good indicator that they don't understand what models are for and what they are not for.

5 comments:

  1. See Lynn Margulis as an example of someone who vehemently objects to this approach in biology. Not that I agree with her.

    ReplyDelete
  2. It goes both ways. I wouldn't assume that everyone who criticizes the use of optimization doesn't know what they're talking about - although it's clear that many don't. As with most things, there are legitimate and illegitimate arguments against the use of optimization; it's really contextual.

    Here's a different version of your last sentence:

    "When people get upset about the people that get upset about models that use optimization, it's a pretty good indicator that they (sometimes) don't understand what those people are upset about!"

    ReplyDelete
  3. Friedman (1953), The Methodology of Positive Economics.

    Optimizing tree leaves FTW.

    ReplyDelete
    Replies
    1. What is amazing about that book is that it forwards a philosophy of science that was already dead at the time the book was written! Economists have a knack for this: just as it became clear to philosophers that Popper's program had failed to demarcate science from non-science, we saw all sorts of economists embracing Popper.

      Delete
    2. Oh, I'm not disputing that. (I've just been flipping through McCloskey's "The Rhetoric of Economics" again. While she certainly wasn't the first, she is one person who conveys these points very well.)

      Still, if we're going to talk optimizing plants and economists, then a nod to Friedman seems pretty appropriate to me!

      Delete

All anonymous comments will be deleted. Consistent pseudonyms are fine.