8 - Analyzing Agent Based Models Flashcards
1
Q
When do you think about measurements ?
A
- think of what you want to measure and analyze before you build your model
- should be clearly linked with research question or goal
2
Q
Why do we need multiple runs?
A
- even though it is a simulation we still have randomness in the sample
- Sampling error > we want to be confident that our results are robust
3
Q
What effects the number of runs?
A
- Number of parameters
- Convention
- Variance stability
- Power analysis (detecting effects)
4
Q
How can you determine the numbers of run you need?
A
- Variance stability
- Power analysis
5
Q
How do you conduct variance stability?
A
- Look at variance across sample sizes (runs)
- Set threshold
- take outcome measure - look at how much variance there is overtime
- set runnumber
6
Q
What are the prerequisites for power analysis?
A
Need to know what kind of analysis you plan to do and what size effect you might have (or a range)
7
Q
What is parameter sweeping/sensitivity analysis?
A
- systematically checking the plausible range of parameter values
- assess the importance of parameters based on model behaviour
8
Q
What are the types of data to analyze from ABMs?
A
- statistical
- graphical
- network-based
- spatial
9
Q
What is model validation?
A
Determine whether the model corresponds to the real-world phenomena
10
Q
What is model verification?
A
Determine whether implemented model corresponds to target conceptual model
11
Q
What is model replication?
A
Someone else implements the model and sees if the results are consistent
