Probabilistic household forecasts based on register data: the case of Denmark and Finland
Solveig Christiansen, University of Oslo
Nico Keilman, University of Oslo
The purpose of this paper is to compute probabilistic household forecasts for Finland and Denmark. We improve on the random share approach of Alho and Keilman by taking advantage of high quality data from population registers and housing registers of Denmark and Finland. Both countries have register data covering the whole populations dating back to the 1980s. We have constructed time series models for household parameters and analysed the empirical prediction errors in those time series models. This allowed us to give an assessment of the expected errors in the household forecasts for the two countries. We forecast, at a 30 year horizon, the number of people occupying the following household positions: dependent child, living with a spouse, living in a consensual union, living alone, lone mother or father, and living in other private household. In addition, the elderly can live in institutional households. Based on probabilistic forecasts of persons broken down by household position, we are able to compute probabilistic forecasts of private households of the following types: one-person households for men and women, cohabiting couples, married couples, and lone fathers and lone mothers.
Presented in Session 110: Issues in stochastic forecasting