Notes:
Idiosyncratic coordination explains how small "errors" (sub-optimal behaviors) can multiply as agents coordinate. This is the most likely reason why so few countries continue to officially use a system that is inefficient and widely unpopular. So which one of the other two options is also a good fit? [See 12.5, "Coordination and Consistency"]
Path length from node A to node B denotes the minimal number of edges that must be traversed to get from node A to node B. The average path length of a network is the average path length between all pairs of nodes. So which of the options could have an average path length of about 4? The EU countries network is a good fit - here, the path length denotes that you must traverse an average of four countries to get from any one EU country to any other. We can't rule out loans or friend networks either, as those are not defined in much depth and could have this average path length. We can't rule out the flight network either, because 4 seems a reasonable path length, depending on the airline.
Next is degree. Degree is the number of edges connected to a node. The degree of a network is the average degree across all nodes. A high degree network is one in which nodes are very interconnected, and a low degree network is one in which there are few connections between nodes. Which of the options are likely to have an average degree of about 2.5? Airline network is out, because we expect almost any airport to have flights to more than 2 or 3 other airports. Friendship network is out as well because most people have more than 2 or 3 friends. In terms of a lending network, it seems reasonable enough that people have loaned to or been loaned from an average of 2.5 others. This average degree also works for the EU geographic network, because it would indicate that countries in the EU, on average, border 2.5 other countries.
Finally, Clustering Coefficient is the percentage of triples of nodes that have edges between all three nodes. In other words, it's the percentage of all possible triangles in a network that are realized. Which one of the two remaining options is likely to have a mid-range clustering coefficient?
The replicator equation tells us that the probability of being type i at time t+1 is equal to the weight of type i at time t divided by the total weight of all types at time t.
Make an educated guess of the fitness levels (given the options) and see if they fit.
After you've plugged in the correct numbers from October, the proportions should match those given for November.
What are the three classes of outcomes - other than equilibrium - that a model can produce?
Complex, periodic, chaotic
2013-12-31 (120min) Coursera open course: Model Thinking by Scott E. Page Start from 10.07.2013, will end on 12.16.2013 Link: https://class.coursera.org/modelthinking-005/lecture/index;
Notes: Idiosyncratic coordination explains how small "errors" (sub-optimal behaviors) can multiply as agents coordinate. This is the most likely reason why so few countries continue to officially use a system that is inefficient and widely unpopular. So which one of the other two options is also a good fit? [See 12.5, "Coordination and Consistency"]
Path length from node A to node B denotes the minimal number of edges that must be traversed to get from node A to node B. The average path length of a network is the average path length between all pairs of nodes. So which of the options could have an average path length of about 4? The EU countries network is a good fit - here, the path length denotes that you must traverse an average of four countries to get from any one EU country to any other. We can't rule out loans or friend networks either, as those are not defined in much depth and could have this average path length. We can't rule out the flight network either, because 4 seems a reasonable path length, depending on the airline. Next is degree. Degree is the number of edges connected to a node. The degree of a network is the average degree across all nodes. A high degree network is one in which nodes are very interconnected, and a low degree network is one in which there are few connections between nodes. Which of the options are likely to have an average degree of about 2.5? Airline network is out, because we expect almost any airport to have flights to more than 2 or 3 other airports. Friendship network is out as well because most people have more than 2 or 3 friends. In terms of a lending network, it seems reasonable enough that people have loaned to or been loaned from an average of 2.5 others. This average degree also works for the EU geographic network, because it would indicate that countries in the EU, on average, border 2.5 other countries. Finally, Clustering Coefficient is the percentage of triples of nodes that have edges between all three nodes. In other words, it's the percentage of all possible triangles in a network that are realized. Which one of the two remaining options is likely to have a mid-range clustering coefficient?
The replicator equation tells us that the probability of being type i at time t+1 is equal to the weight of type i at time t divided by the total weight of all types at time t. Make an educated guess of the fitness levels (given the options) and see if they fit. After you've plugged in the correct numbers from October, the proportions should match those given for November.
What are the three classes of outcomes - other than equilibrium - that a model can produce? Complex, periodic, chaotic