Validating a dynamic population microsimulation model: Recent experience in Australia

Abstract

Available published research on microsimulation tends to focus on the results of policy simulations rather than upon validation of the models and their outputs. Dynamic population microsimulation models, which age an entire population through time for some decades, create particular validation challenges. This article outlines some of the issues that arise when attempting to validate dynamic population models, including changing behaviour, the need to align results with other aggregate ‘official’ projections, data quality and useability. Drawing on recent experience with the construction of the new Australian Population and Policy Simulation Model (APPSIM), the article discusses the techniques being used to validate this new dynamic population microsimulation model.