64th ISI World Statistics Congress - Ottawa, Canada

64th ISI World Statistics Congress - Ottawa, Canada

Projection of multiple demographic characteristics and policy relevance: the Demosim microsimulation model at Statistics Canada

Author

SV
Samuel Vézina

Co-author

  • J
    Jean-Dominique Morency
  • P
    Patrice Dion

Conference

64th ISI World Statistics Congress - Ottawa, Canada

Format: IPS Abstract

Session: IPS 395 - From data to policy via modern microsimulation approaches

Monday 17 July 2 p.m. - 3:40 p.m. (Canada/Eastern)

Abstract

Cohort-component projection models typically used by national statistical agencies (including Statistics Canada) are not tailored for demographic projections that comprise multiple characteristics beyond age, sex and region. There is a need, however, for projections that can simultaneously account for key ethnocultural characteristics (e.g., Indigenous identity, immigrant status, racialized group) and socioeconomic variables (e.g., education and labour force participation). Free from the matrix infrastructure inherent to conventional projection methods, microsimulation models provide the flexibility to produce such disaggregated population projections. Demosim is an ever-evolving demographic projection microsimulation model that has been developed and maintained at Statistics Canada for more than two decades. This model takes as its starting population the Canadian Census of Population (detailed questionnaire) microdata file and projects each individual’s life course trajectory by simulating key events including fertility, mortality, geographic mobility and changes in education level. The power of Demosim resides in its capacity to provide detailed projections of the population and to account for population heterogeneity that may affect aggregate results. In turn, to estimation of the large number of conditional transition probabilities often requires extensive analysis and the use of multiple datasets such as: censuses, surveys, administrative data, and linkages between datasets. This presentation provides an overview of Demosim and its key statistical foundations. Furthermore, through several concrete examples of studies that used Demosim, we demonstrate the power and versatility of Demosim to answer a variety of research questions, thanks to its capacity to project a large number of characteristics and to mimic complex demographic processes. We show how Demosim results have contributed, for instance, to providing the public and policy makers with detailed projections that reflect the future ethnocultural diversity of the population or focus on the Indigenous populations in Canada.