GLOSSARY

A shared vocabulary for an interdisciplinary field

Core terms include abbreviations, full names, and the working definitions used throughout AI × ECON.

01

ACE

Agent-Based Computational Economics
Agent-Based Computational Economics

A bottom-up approach that generates and analyzes aggregate economic dynamics from heterogeneous agents and their local interactions.

02

ABM

Agent-Based Model
Agent-Based Model

A computational model that explicitly encodes autonomous agents, behavioral rules, environments, and interactions.

03

DSGE

Dynamic Stochastic General Equilibrium
Dynamic Stochastic General Equilibrium

A structural macroeconomic model combining intertemporal optimization, stochastic shocks, expectations, and market clearing.

04

HANK

Heterogeneous-Agent New Keynesian
Heterogeneous-Agent New Keynesian

A macroeconomic framework combining household heterogeneity and asset distributions with New Keynesian nominal rigidities.

05

DML

Double / Debiased Machine Learning
Double / Debiased Machine Learning

A method for estimating target causal parameters with high-dimensional nuisance functions using orthogonal scores and cross-fitting.

06

MARL

Multi-Agent Reinforcement Learning
Multi-Agent Reinforcement Learning

A reinforcement-learning paradigm in which multiple learning agents interact, compete, or cooperate in a shared environment.

07

Synthetic Subjects

Synthetic Subjects
合成被试

Model-generated computational subjects conditioned on experimental instructions and agent characteristics; they are not equivalent to real human samples.

08

Economic World Model

Economic World Model
经济世界模型

An internal model of economic states and their dynamics used to simulate constrained multi-agent interaction and support planning, forecasting, and scenario analysis.

09

Mechanism Identification

Mechanism Identification
机制识别

The task of distinguishing alternative micro-level behaviors and institutional mechanisms that can generate similar aggregate outcomes.

10

Out-of-Sample Validation

Out-of-Sample Validation
样本外验证

Evaluation on data that were not used for training, calibration, or model selection.