Storing work from ABM study from Intro to ABM via Santa Fe Institute.
Below, my work / progress through the study is documented, newest project to oldest from the top.
Extention of the El Farol model to build rewards system and monitoring around the model.
Part of NetLogo training the task was to create a simulation to see what happens in a simple job market.
- The x position of each agent represents their wealth 0-100
- Each tick, agents are deducted living expenses
- Each tick, agent earns income if they have a job
- Agents lose or find jobs on their own at rates set by job-loss-rate and job-find-rate sliders
- If an agent has a job for more than 5 ticks, they create a friend (who is also employed). The rate of friendships being created this way is controlled by a slider that one can adjust
- Friendships are lost over time
- If agent is unemployed, it can ask a current friend that has high wealth for a job (at the same rate as their friend). The rate of them getting a job this way is set by a slider that one can adjust
- Agents change color based on being poor (less than 10 wealth), being wealth (more than 90), having a job, or being unemployed
- Social connections are labeled with links
Recreating the Heroes And Cowards Activity as an ABM
- Each agent is either a hero or a coward
- Each agent selects one other agent to be his friend, and another agent to be his enemy
- If an agent is a hero, he will attempt to get between his enemy and his friend
- If agent is a coward, he will attempt to put his friend between himself and his enemy
Warming up, creating a simple model to understand wealth distribution.
- Created simple economy model and charting
- Analyzed results with BehaviorSpace
Simple tutorial following netlogo default tutorial
- agents move randomly
- if there's grass, agent eats it to get energy
- agent has chance of reproducing.
- if patch has no grass, it has a chance of growing back