2020/06/15 by Amarin Siripanich, Siripanich, Amarin, Taha Hossein Rashidi +1
Decision Sciences · Engineering · Social Sciences · #Applications (stat.AP) #FOS: Computer and information sciences #FOS: Economics and business #FOS: Physical sciences #General Economics (econ.GN) #Insurance, Mortality, Demography, Risk Management #Migration, Aging, and Tourism Studies #Physics and Society (physics.soc-ph) #Transportation and Mobility Innovations #demographic modeling and climate adaptation
paper · pdf · doi:10.48550/arxiv.2006.09474
openalex publication_date 2020/06/15 · openalex created_date 2022/09/14 · openalex updated_date 2026/07/28
Many dynamic microsimulation models have shown their ability to reasonably\nproject detailed population and households using non-data based household\nformation and dissolution rules. Although, those rules allow modellers to\nsimplify changes in the household construction, they typically fall short in\nreplicating household projections or if applied retrospectively the observed\nhousehold numbers. Consequently, such models with biased estimation for\nhousehold size and other household related attributes lose their usefulness in\napplications that are sensitive to household size, such as in travel demand and\nhousing demand modelling. Nonetheless, these demographic microsimulation models\nwith their associated shortcomings have been commonly used to assess various\nplanning policies which can result in misleading judgements. In this paper, we\ncontribute to the literature of population microsimulation by introducing a\nfully integrated system of models for different life event where a household\nalignment method adjusts household size distribution to closely align with any\ngiven target distribution. Furthermore, some demographic events that are\ngenerally difficult to model, such as incorporating immigrant families into a\npopulation, can be included. We illustrated an example of the household\nalignment method and put it to test in a dynamic microsimulation model that we\ndeveloped using dymiumCore, a general-purpose microsimulation toolkit in R, to\nshow potential improvements and weaknesses of the method. The implementation of\nthis model has been made publicly available on GitHub.\n