ABSTRACT

One of the important issues relevant to the spatial microsimulation modeling approach of small area estimation is the difficulty in validating the model outputs and measuring the statistical reliability of the estimates. This is because the microsimulation modeling technology (MMT) approach simulates a spatially disaggregated synthetic micropopulation data set that previously did not exist and uses this data set to produce the small area estimates. Validation processes in MMT typically check the model outputs in a range of systematic ways to reveal errors in the outputs. Without examining such errors from the model, a modeler or researcher cannot claim the accuracy of the estimates produced by the model. Measures of statistical reliability can also be used to assess the stability of the model estimates.