The notion of a parameterized ABM would extend our existing ABMs with a namespace of values as well as some means of accessing those values in transition rates, model initializaiton, reporting processes, etc. The details of this need to be worked out.
There should be a clear distinction between model parameters (exogenous) vs endogenous values, even though there is some degree of subjectivity as to what should be a parameter.
Parameters should be able to take continuous values, categorical values, discrete ordinal values, etc.
We could refer to a namespace of values as a "scenario". We often want to run the same model under different scenarios in a systematic way.
We would want to be able to compose models in a way that permits automatic migration / composition of the parameters of each model.
The notion of a parameterized ABM would extend our existing ABMs with a namespace of values as well as some means of accessing those values in transition rates, model initializaiton, reporting processes, etc. The details of this need to be worked out.
There should be a clear distinction between model parameters (exogenous) vs endogenous values, even though there is some degree of subjectivity as to what should be a parameter.
Parameters should be able to take continuous values, categorical values, discrete ordinal values, etc.
We could refer to a namespace of values as a "scenario". We often want to run the same model under different scenarios in a systematic way.
We would want to be able to compose models in a way that permits automatic migration / composition of the parameters of each model.