Simulating the Characteristics of Populations at the Small Area Level: New Validation Techniques for a Spatial Microsimulation Model in Australia.
- Published: January 2013
- Authors: Azizur Rahman, Ann Harding, Robert Tanton and Shuangzhe Liu
- Journal Title: Computational Statistics and Data Analysis
- Journal Volume: 57
- Number: 1
- Article Pages: 149-165
- Journal ISSN: 0167-9473
- Research Area: Microsimulation
- Comments: Originally pub online 25 June 2012.
- Keywords: Geographic map, housing stress, Reweighting, small area estimation, spatial microdata and Statistical significance test
These days spatial microsimulation modelling plays a vital role in policy analysis for small areas. Most developed countries are using these tools in ways to make knowledgeable decisions on major policy issues at local levels. However, building an appropriate model is very difficult for many reasons. For example, the creation of reliable spatial microdata is still challenging. In addition there has not been much research on testing statistical significance of the model outputs yet, and deriving estimates of how reliable these outputs may be. This paper deals with the spatial microsimulation model building procedure for simulating synthetic spatial microdata, and then estimating small area housing stress in Australia. Geographic maps for small area housing stress estimates are illustrated. The research also demonstrates a new system to test the statistical significance of the model estimates.