To help model financial systems, some economists are turning to biology
theatlantic.com
theatlantic.com
If you want to see what economic understanding will eventually look like, just cut to the chase and read The Nature of Economies today.
It's so damn close to a theoretical model and it exists in real life! Data is easily accessed and interpreted. Compare this to the markets for real estate, used cars, or online advertising. It's all precise enough to tie to monetary rules, central bank actions ...the policy levers top economists have available in a normal year.
Whatever your theory, it will be easier to test for publicly traded stocks or commodities.
The problem is that public markets are formalized, centralized markets. They are rational by design. This is unlike every other market.
To extend their ecological metaphor a different way... Researching public markets is like cells in a petri dish, not a real ecology.
Last year, I was reading Proudhon's criticism of the speculative part of financial markets. His main objection was that speculation was adding absolutely no value to the world, by opposition to investors (who help making something useful real) and merchants (who displace products from somewhere where they're abundant to somewhere where they're scarce). It hitted me back then: actually, they do produce something, they produce data. Financial markets have produced the longest lived and most detailed dataset ever (I can't think of a bigger one, at least). Would be a waste not to try to use it.
Evolutionary economics has been a thing since 1898 https://en.wikipedia.org/wiki/Evolutionary_economics
And had a resurgance in the 1990s, see for example R. H. Day & P. Chen, eds, ‘Nonlinear dynamics and evolutionary economics’, Oxford University Press, USA