Thread by @ole_b_peters

1/5 EE will significantly transform economic theory. It's only a matter of time (pun very much intended).

EE tells you to look at the ergodic properties of your system: include time. Say the system is a collection of individuals whose wealths grow multiplicatively with noise.

https://twitter.com/ashtondalelove/status/1134046246090547200

2/5 Then the ensemble average grows at a rate that's higher than the time-average growth rate. This implies increasing inequality in the ensemble: some very rich individuals must skew the average.

3/5 Conceptualize this through EE, and you'll find a nice measure of inequality: the time-integrated difference between the growth rate of the expectation value and the time-average growth rate. That's how much inequality has accumulated over time, a very natural measure.

4/5 Then look at the literature. Henri Theil discovered the same measure, which he called "mean log deviation," in 1967, saying it was "against intuition." Amartya Sen called Theil's work "not exactly overflowing with intuitive sense" - no one knew what it meant.

Now we know.

5/5 There are many other examples from other branches of economics.
We wrote this one up for @RoyalStatSoc. The paper got a bit messy following requests by reviewers but the key idea is there.
rss.onlinelibrary.wiley.com/doi/epdf/10.11…