Stochastic techniques for estimating uncertainty intervals for demographic forecasts are still not widely used by statistical institutes. One reason for this may be the complexity of the Monte Carlo approach, which requires the calculation of a thousand or more variants of the deterministic forecast. This paper discusses a technique to estimate forecast intervals for demographic indicators from only 6 variants.
The method was developed for the 2016-based regional population and household forecast of Statistics Netherlands and the Netherlands environmental assessment agency. This paper focuses on the application to a national population forecast, where the results can be compared to those from a stochastic forecast.
