Agent-Based Modeling
Routing Summary
This topic covers agent-based modeling theory, consumer behavior models, social dynamics, calibration/validation methods, and applications. Contains 5 sub-topics and 34 total concept notes derived from 4 papers.
- For ABM theory (definitions, emergence, heterogeneity, decision rules) → Foundations
- For consumer ABM models (Karakaya utility model, CUBES behavioral model) → Consumer Behavior
- For social interaction mechanisms (WOM, networks, diffusion, market dynamics) → Social Dynamics
- For calibration and validation (GA, HM+ABC, uncertainty quantification, case studies) → Calibration and Validation
- For ABM application domains (marketing, flows, finance, organizations, risk) → Applications
- For LLM-powered agents (generative agents, silicon samples / simulated respondents, homo silicus, validity and calibration of LLM populations) → LLM-Powered Agents
Sub-topics
- Foundations — COVERS: ABM methodology and principles, emergent phenomena, heterogeneity, ABM vs equation-based modeling, agent decision rules and bounded rationality
- Consumer Behavior — COVERS: Karakaya 4-component utility model, logit purchase decisions, CUBES behavioral attitudes (mistrust, opportunism, conditioning, innovativeness, imitation), behavioral primitives with threshold activation
- Social Dynamics — COVERS: WOM mechanisms (positive/negative), opinion leaders (exogenous vs emergent), social network formation, product adoption/diffusion models, network topology effects, market share equilibrium and lock-in
- Calibration and Validation — COVERS: ABM calibration overview (3 approaches), GA calibration (chromosome encoding, GA operators, convergence), RAM fitness evaluation (macro/micro), HM+ABC framework (implausibility score, wave-based pruning, rejection sampling ABC), uncertainty quantification (model discrepancy , ensemble variance , observation uncertainty ), case studies (SugarScape, territorial birds vs GA/SA/EA, RISC Scottish farms POM), experimental design (LHS, parameter sensitivity), validation (plausibility standard)
- Applications — COVERS: marketing strategy (4Ps under WOM), flow simulation (evacuation, traffic), financial market simulation, organizational simulation, operational risk modeling
- LLM-Powered Agents — COVERS: the LLM-agent paradigm for ABM (Gao et al. 2023 survey), generative-agent architecture with memory/reflection/planning (Park et al. 2023), silicon samples and algorithmic fidelity (Argyle et al. 2022), homo silicus economic experiments (Horton 2023), persona-mixture calibration, LLM vs rule-based agents, opinion-alignment metrics (Santurkar et al. 2023), and a validity/bias threat taxonomy with a validation protocol (8 notes, added 2026-09-18)
Key Concept Dependencies
ABM Methodology -> Emergent Phenomena -> ABM vs Equations
| |
v v
Heterogeneity Decision Rules
| |
v v
Consumer Models Social Dynamics
| |
v v
Calibration <------> Validation
/ \
GA HM+ABC
|
Uncertainty QU
|
v
Applications
Source Papers
| Paper | Authors | Year | Focus | Notes Produced |
|---|---|---|---|---|
| abm_word_of_mouth | Bonabeau | 2002 | ABM methodology, applications, adoption model | Foundations + Applications + Social Dynamics |
| abm_consumer | Karakaya, Badur & Aytekin | 2011 | Marketing strategies with WOM | Consumer Behavior (Karakaya) + Social Dynamics |
| abm_human_behaviour | Ben Said, Bouron & Drogoul | 2002 | CUBES consumer behavior simulator | Consumer Behavior (CUBES) + Calibration (GA/RAM) |
| calibration_ABM | McCulloch, Ge, Ward, Heppenstall, Polhill & Malleson | 2022 | HM+ABC calibration with uncertainty quantification | Calibration Methods (HM+ABC framework, HM, ABC, UQ, case studies) |
Recent Ingestion Log
| Date | Files | Notes Created | Notes Updated |
|---|---|---|---|
| 2026-04-11 | calibration_ABM.pdf | 5 concept notes (HM+ABC framework, History Matching, ABC, UQ, Case Studies) | Calibration and Validation _Index, Calibration Methods _Index |
| 2026-04-10 | 3 PDFs (abm_consumer, abm_human_behaviour, abm_word_of_mouth) | 29 concept notes, 16 index files | — |