Population Change and Lifecourse

A Planning Workshop on Micro-Simulation

Life Course, Population Change, and Micro-Simulation: A Planning Workshop

25 October 2010

Cluster members, representative of partner agencies, graduate students and guests from Europe were the participants in a one-day workshop of the Population Change and Lifecourse Strategic Knowledge Cluster. Its aim is to plan a program to promote micro-simulation for research, particularly studies with policy relevance.

The workshop schedule included a discussion of the current state and the future of microsimulation programs in Canada and Europe, research in Canada using micro-simulation programs, the teaching of courses on microsimulation, and policy-related research programs involving micro-simulation. The participants also talked about the role of the Cluster in promoting the use of micro-simulation in research, training, and knowledge mobilization; and, the next steps that the Cluster should undertake to promote microsimulation.

Microsimulation in Statistics Canada

Statististics Canada has developed Modgen, a generic microsimulation programming language, which supports the creation, the development and the maintenance of microsimulation models. Some of the microsimualtion models created using the Modgen technology are:

  • LifePaths, a dynamic longitudinal microsimulation model of individuals and families. Using behavioural equations estimated using a variety of historical micro-data sources, LifePaths creates statistically representative samples consisting of complete lifetimes of individuals.
  • Demosim is a microsimulation model designed for specific population projections. Demosim allows to project a large number of characteristics of the population all at the same time, while also taking into account differentials in demographic behaviours between sub-groups of the population.
  • RiskPaths is a simple demographic competing risk microsimulation model for the study of changes in fertility and partnership patterns. RiskPaths permits analysis of the contribution of changes in single processes on aggregate measures like cohort fertility or childlessness.

In Europe, several microsimulation models are being used for diverse applications including population forecasting (for example, MICMAC), tax-benefit modelling (EUROMOD), pensions and pension reforms scenarios (DESTINIE, MOSART), and simulating social policy in an aging society (SAGE).

Use of Microsimulation for Policy-Relevant Research

Partners of the Population Change and Lifecourse Cluster have keen interest on the use of microsimulation. At the Public Health Agency of Canada, for example, some of the projectss that could benefit from microsimulation modeling include:

  • Health Status and Health Costs by Income Level in Canada
  • Costs of Childhood and Adult Obesity
  • Long Range Scenario Planning on Health Literacy in Canada.

Similarly, there are projects at Citizenship and Immigration Canada which could benefit from microsimulation modelling.

Two research projects of Cluster members, funded by the Human Resources and Skills Development Canada, are using microsimulation:

  • Monetizing the costs of care-related labour market interruptions by Janet Fast, Donna Dosman, Yann Décarie, Martin Spielauer
  • Caregiver Supply and Demand: Preparing for the Future Needs of Older Canadians by Janice Keefe, Jacques Légaré, Bonnie-Jeanne MacDonald, and Michel Grignon

A conceptual framework that could link population change, life courses, and policies is the Olivia Framework. As Peter Hicks, the proponent of the framework, notes, "it has the potential to integrate life course analysis into a broader approach to social policy analysis -- especially in supporting needed data collection and the development of integrative analytic tools such as microsimulation tools".

Training Students on Microsimulation

Apart from Statistics Canada that offers workshops on microsimulation, currently there are two courses on microsimulation offered in Canada: at the Glendon Department of Economics with Arthur Younger as professor; and at the Institut national de la recherche scientifique with the course Perspectives et micro-simulation taught by Alain Bélanger. As noted by Alain Bélanger, the two main challenges in teaching microsimulation are the few number of students interested in taking the course, some of whom are not ready in terms of their knowledge of computer and computer programming, and the lack of pedagogical materials

Some Activities that the Cluster could undertake to promote microsimulation

  • Create a Canadian microsimulation portal where a community of users could share ideas, problems, solutions, and codes.
  • Support various ways of training microsimulation at different levels of users and programmers. These trainings could be workshops, summer institutes with specific content (such as aging, health, retirement income, etc), or courses in the universities.
  • Support the development of pedagogical materials.
  • Support applications for funding research and training on microsimulation, for example, the Network for Centre of Excellence on Health in an Aging Population, or SSHRC Partnership Grant.

Statistics Canada Workshop: From Traditional Demographic Calculations to Projections by Microsimulations

Ottawa, October 26, 2010

The workshop was attended by a number of participants from the Cluster, including graduate students who benefited from the 2010 special student competion. It was conducted by André Cyr, Julien Bérard-Chagnon, Éric Caron Malenfant, and Dominic Grenier from Statistics Canada. Aimed at demystifying the estimation and demographic projection methods used at Statistics Canada, the workshop had two parts:

  • The production of demographic estimates. These estimates require the simultaneous use of multiple sources of data with several challenges. They are based on censuses, which are adjusted via reverse record check. Administrative sources are used to estimate the population and its demographic components.
  • The two main methods for demographic projections. The first method, by component, uses aggregate data to predict the effective future of the population according to different hypotheses regarding its evolution. The second method, by microsimulation, uses the censuses to predict individuals one by one based on their characteristics and a series of hypotheses. Various surveys and administrative sources are used to calculate the risks of simulated events.