5 - Creating Agent-Based Models

Question 1

What is a model?

Answer 1

• a conceptual/textual description
• a software-based implementation

Question 2

What are the 2 major categories of ABMs?

Answer 2

• Phenomena-based modelling
• Exploratory modelling

Question 3

What is phenomena-based modelling?

Answer 3

• model captures the referent patterns
• e.g. segregation, spiral formation, oscillations populations of agents

Question 4

What is exploratory modeling?

Answer 4

• Create agents, define their behaviour and explore the patterns that emerge

Question 5

What is the top-down approach of conceptualizing models?

Answer 5

• develop the conceptual model first then implement it
• need to have the research question, design agents and their rules of behaviour, elements of the environment
• refine and revise until it has enough detail to be coded

Question 6

What is the bottom-up approach of conceptualizing models?

Answer 6

• choose a domain of interest, start coding something relevant to the domain, adding in the conceptualizations, mechanisms, properties etc.

Question 7

How does the conceptualization of models work in practice?

Answer 7

It is a combination of top-down and bottom-up approach

Question 8

What is the AMB design principle?

Answer 8

Start simple and build towards the question you want to answer
> start with the simplest set of agents and relevant behaviour
> always keep your research question in mind

Question 9

What is the purpose of verification in ABM?

Answer 9

Ensuring that a computational model faithfully implements its target conceptual model

Question 10

True of False: Simpler models are easier to verify and scale up from

Answer 10

True

Question 11

Based on what factors do you choose your agents?

Answer 11

• Agent properties (preferences)
• Environmental characteristics and stationary agents
• Agent behaviour
• Time steps
• Parameters
• Measures

Question 12

How do you examine a model?

Answer 12

• Multiple runs (keep time constant but random seeds)
• Parameter sweeping and results collection (robustness and sensitivity analysis)
• Data Analysis (statistical inference, visualization, etc.)

Question 13

What is the first step of building a model?

Answer 13

Formulating the driving research question