How does our initial "seed" model work?Agents and ResourcesWe assume a world containing a number of agents and a number of types of resource. * Each agent has a capacity to get or create a certain amount of each resource. We call this the ResourceGettingCapacity? or RGC. * Each agent has a consumption rate for each resource. We call this the ResourceConsumptionRate? or RCR. * Each agent has a minimal need for each resource. We call this the "need". Typically, if the agent falls below this need in any resource, the agent dies. * And at any time, the agent has a current store of each resource. We call this "current". For example, in a world with only two kinds of resource : food and drink. Any possible agent will have the following characteristic : |resource name | need | rgc | rcr | current |food | 4 | 2 | 2 | 4 |drink | 3 | 3 | 2 | 6 Each time-step, this agent will acquire more of the resource according to its RGC for that resource, and will consume as much as its RCR. So this agent will get 2 units of food and consume 2 units of food. Its store of food remains constant. For drink, it gets 3 units but only consumes 2. Hence, this time step its overall current store will have increased by 1. Result : happiness. On the other hand, an agent which can not at least break even in all resources will die when its current store falls below its need. Networks and EconomiesAgents are placed within a social network. That is, each has a neighbourhood of a number of local other agents. Each time-step the simulation runs, an agent will get and consume resources, but can also choose to interact in some way with the other agents in its neighbourhood. Different kinds of agents allow different kinds of economy to be examined, agent who give gifts, agent who barter or try to barter for a profit. Soon, we'll have agents that work with different kinds of money. Because we also consider the behaviour of the economy may be affected by the underlying social structure, the social network itself can be configured in different ways. We can set up locally connected, randomly connected and fully connected. Soon we'll add SmallWorlds (locally connected with some random re-wiring) and ScaleFreeNetworks. And eventually we expect to be able to look at economies who's network is itself rewired by the economic activity. ObservationsWe're interested in both how the agents succeed individually and the performance of the whole economy. We can measure a number of statistics about the overall state of the economy. Its overall surplus, its overall production and the distribution of wealth within the society. (sDev) Missing Features ===Opportunity Cost ==== Typically in economics getting any given resource is assumed to have some kind of opportunity cost. In other words, if I decide to get 3 bannana's then, I will not have time to get 4 coconuts. This model has no concept of opportunity cost at the individual level (there is some opportunity cost at the societial level since a unit of food could either be stored, or go to feed some individual). This will pretty much make it so that there is no benifit to not producing some resource. However in real life and most economic models, there is at least the advantage of gaining more leisure time. Without opportunity cost on the indivdual level this model will likely be considered inaddaquate for investigating the type of questions proposed for it by the vast majority of economists. : Good point. I'll have a think about this. Cheers. -- PhilJones |
Can't locate object method "startform" via package "CGI" at optimaes.cgi line 1479.
For help, please send mail to the webmaster ([no address given]), giving this error message and the time and date of the error.
![[Home]](./ts.jpg)