Forecasting social inequality using agent-based modelling
summary
Agent‑based model of UK demography uses ONS and census inputs and Python (numpy, pandas) to run 150‑year simulations on 10,000 agents across many AWS cores. Comparing descendants trusts versus direct inheritance shows a sustained drop in inequality measured by the Gini coefficient for up to a century, earlier wealth injections for grandchildren, but a counterintuitive rise in parent–child wealth correlation. Model assumptions include 5% saving, 3% returns, retirement annuity and simple coupling/fertility rules.
AI-written summary · about
chapters (9)
- 0:00Agent-based modelling intro
- 7:53What is an agent-based model
- 10:20Data for inheritance model
- 14:12Building agent rules
- 21:00Testing inheritance rules
- 24:12Agent wealth trajectories
- 27:27Measuring inequality with Gini
- 32:48Simulation results
- 33:53Q&A on model
related talks
31mCounterfactual Analysis with Bayesian Models
15mBeyond Efficiency
35mComputer Simulations of Social Behavior With Cellular Automata
52mAn Evolutionary Model of Economic Behavior, Bounded Rationality, and Intelligence
37mDid Your Rollout Actually Work?
37mAI for Good- Then and Now
24mState of the Art Open Source Software for AI Fairnessevaluation
26mParadoxes in Data Science