A Comparison of Data Merging Methodologies for Extending a Microsimulation Model

This paper compares the use of two different methodologies for merging data from different sources in developing or extending a microsimulation model. The first uses a relatively simple means-based methodology, and the second uses regression. The advantages and disadvantages of each are discussed and their ability to accurately retain the information in the original data sources is tested through their application in the development of a model of child care services and subsidies.