Statistical Implementations of Agent‐Based Demographic Models

摘要: A variety of demographic statistical models exist for studying population dynamics when individuals can be tracked over time. In cases where data are missing due to imperfect detection individuals, the associated measurement error accommodated under certain study designs (e.g. those that involve multiple surveys or replication). However, interaction and underlying dynamic process complicate implementation agent-based (ABMs) demography. a Bayesian setting, traditional computational algorithms fitting hierarchical prohibitively cumbersome construct. Thus, we discuss approaches ABMs demonstrate how use multi-stage recursive computing emulators fit in such way alleviates need have analytical knowledge ABM likelihood. Using two examples, model survival compartment COVID-19, illustrate procedures implementing ABMs. The describe intuitive accessible practitioners parallelised easily additional efficiency.