Economic forecasting with Agent based model under Bayesian framework

This project is about implementing a technique called Agent-Based modelling (ABM) so it can work better in real-world application. Particularly, it aims to help policy makers to do more adaptive decisions when the whole economics environment changes. For example, how to set the federal interest rate after COVID-19 panic? This model could simulate how all kinds of people, regulators, corporations, banks, or investors interact with others and how that interaction could cause specific things to happen to them and to the market more broadly. Even ABM can provide a realistic view of economics system, the validation and calibration are undeniably complex, which leads the result is hardly to interpret. Moreover, it is highly possible to see one model may performance well in one period and other competitor models may do better in another period, but there is no obviously criteria for choosing which model is the most realistic.
Second, the computational cost of ABM is essentially large. These issues have been a barrier to their more widespread adoption in economics application.

Our research will focus on solving those two problems. First, we would like to apply a Bayesian approach so we could compare models systemically.

Faculty Supervisor:

Matheus Grasselli;Pengfei Li

Student:

Meng Yuan;Yicheng Chen

Partner:

Hummer

Discipline:

Engineering - computer / electrical

Sector:

Finance, insurance and business

University: