Microsimulation models are computer models that operate at the level of the individual behavioural entity, such as a person, family, or firm. Such models simulate large representative populations of these low-level entities in order to draw conclusions that apply to higher levels of aggregation such as an entire country. This type of model is distinct from aggregate models whose explanatory variables already represent collective properties. An example of such an aggregate explanatory variable might be the national unemployment rate. Certain types of modeling problems are best dealt with using microsimulation whereas for others an aggregate approach is more appropriate.
Statistics Canada has developed a number of microsimulation models as well as general purpose tools that assist in their construction. Some of these models and tools can be downloaded from the pages linked below. A variety of analyses, some originating from inside Statistics Canada and some from outside, have been performed using these models.
For more information about microsimulation activities at Statistics Canada, click one or more of the links below.
SPSD/M
Social Policy Simulation Database and Model (SPSD/M) is a detailed cross-sectional microsimulation model of individuals and families. It is based on a non-confidential annual database constructed using a variety of survey and administrative data sources. It is used for policy development and analysis of Federal and Provincial tax and transfer programs, as well as for analyzing issues related to income distribution.
Demosim
Demosim is a microsimulation model designed for population projections, starting with the Canadian census microdata file as the base population. It generates projections of a large number of population characteristics at different geographical levels and according to different scenarios of population growth.
Health Models
This overview outlines the health-related simulation models developed by the Health Analysis Division of Statistics Canada.
CrimPath-1
CrimPath-1 is an R prototype simulation program created at Statistics Canada, in collaboration with the Department of Justice. The prototype is designed to model the trajectories of individuals involved in the criminal justice system (CJS) in Ontario and Saskatchewan and project the impacts of hypothetical justice interventions on CJS outcomes and associated costs. Ultimately, the goal is to build a simulation model that can help identify which program, policy or legislative changes could be the most effective in reducing (re-)contact with the CJS, on a national scale, and in reducing the overrepresentation of Indigenous and racialized people within the system. More information on the prototype and the estimated counterfactual scenarios can be obtained by contacting statcan.ccjcss-ccsjsc.statcan@statcan.gc.ca.