Forecasting Incapacity Benefit

Client: Department for Work and Pensions (DWP)
Student: Daniel Hill
Supervisor: Dr Stephan Onggo (MSc OR Project)

Summary

The aim of this project is to examine uncertainty in the micro-simulation model used to forecast incapacity benefit used in the Department for Work and Pensions (DWP). The micro-simulation model uses the DWP’s simulation system called 'Genesis' which is mainly written in SAS. A number of possible techniques for quantifying forecast error are proposed which range from a historical analysis to a more complex combination of parameter evaluation and sensitivity analysis. Each option is presented with a set of advantages and disadvantages, as well as suggestions for further work. Some wider questions of uncertainty in the incapacity benefit forecast are considered, with a view to stimulating debate around the uncertainty and the nature of forecasting.