A deterministic differential equation epidemic model produces the values for the state variables over the simulation time horizon, which consists of the numbers of individuals in each disease state. A detailed agent-based model produces data on many more variables in a single simulation run, from agent disease states to agent interaction outcomes, to agent behavioral responses activities. The stochastic nature of the simulation also requires many simulation runs to properly characterize uncertainty. Recording all of the data the model produces is prohibitive due to the overhead of recording and storing the data. A subset of “indicator variables” consists of summary statistics computed by the model and logged as the simulation progresses. The indicator variables are initially used for debugging and to determine whether the model is producing reasonable results; later in the modeling process indicator variables are used for model calibra- tion, verification, and validation. Indicator variables include infected and colonized counts at each time, new colonizations due to contacts with infected and colonized individuals, dendograms for selected agents that portray their entire disease state history, and many more.
A deterministic differential equation epidemic model produces the values for the state variables over the simulation time horizon, which consists of the numbers of individuals in each disease state. A detailed agent-based model produces data on many more variables in a single simulation run, from agent disease states to agent interaction outcomes, to agent behavioral responses activities. The stochastic nature of the simulation also requires many simulation runs to properly characterize uncertainty. Recording all of the data the model produces is prohibitive due to the overhead of recording and storing the data. A subset of “indicator variables” consists of summary statistics computed by the model and logged as the simulation progresses. The indicator variables are initially used for debugging and to determine whether the model is producing reasonable results; later in the modeling process indicator variables are used for model calibra- tion, verification, and validation. Indicator variables include infected and colonized counts at each time, new colonizations due to contacts with infected and colonized individuals, dendograms for selected agents that portray their entire disease state history, and many more.
การแปล กรุณารอสักครู่..