24.5.08

modelling the physics of rice

I'd like to know if anybody has modelled the price of rice, for say a country like bangladesh, with both deterministic and non deterministic considerations, with both game theory and maim theory embedded.

Non-deterministic process need to be understood because the world doesn't move in straight lines according to objective rules, especially when human actors with different values are involved. Maim Theory is where participants in an action theory actually have no idea of what their interests are or obey any norms because of a messed up habitus. Extreme Maim Theory describes the situation when escalation/exacerbation is the dominant modus operandi of at least one of the participants.

It would need a lot of reliable empirical data to run historical training type grooves on, like they do to train Climate models...would need a helluva lot of spatial info about how stuff grows. It wouldn't suit a craponomist in a Kafir Development Bank, nor a self pleasuring physicist in some Terry Pratchettesque Unseen University setup, nor their bastard spawn.

transportation (oil link)
spatial ripening fluxes
mobile communications habits(complicated feedback thing)
micro, meso and mass storage hubs
wholesale to retail relations
the proportions of the different varieties
political polarity of human participants
irrigation type (oil link)
fertiliser type (gas link)
consumer link
hype link (media-political link)
seasonality
shock events
foreign flux (yes the outward pumps in chapainawabganj AND the inward 'aid' pumps)

ok, I'm scared now.

2 comments:

Saif said...

This is extremely doable - with the existing tools of the "craponomist's" trade, as you call it. The good amount of this data is likely available as well - though probably dispersed. I wouldn't call it Maim theory. It's straight up empirical work.

fug said...

i know that hussain zillur rahmans outfit, pprc did stuff on this. i wish the same heavy duty brains were dedicated to understanding the human mechanics and relations of this beast... as were employed as quants.

when things get complicated models mess up. unless you have a very funky model.

wonder if BAU do this kinda stuff, or how close they are to it.