divilian / specstar

Combines SPECscape and SPECnet into one project
1 stars 1 forks source link

Compute Gini coefficient wrt proto institutions #50

Open divilian opened 5 years ago

divilian commented 5 years ago

Possibly interesting question: generally speaking, after steady-state, do rich agents end up being in protos together with other rich agents, and poor with poor?

It seems like it would be interesting to compute and analyze the following metrics:

  1. Compare the protos to each other: a. Treat each proto as its account balance, and compute the Gini coefficient. b. Treat each proto as the average of its constituent members' personal wealths, and compute the Gini coefficient.
  2. Look at each proto individually: compute the Gini coefficient of its constituent members.

These get at the question in different ways. The first says "do protos themselves exhibit inequality similar to how agents do?" The second says "within a proto, do its member agents exhibit inequality as they do when all agents are globally compared with one another?"