Abstract: | The paper discusses the problem of modelling demographic variables for the purpose of forecasting. Two empirical model selection procedures are applied to suggest final form forecasting equations for Australian marriage rates. The suggested models are then assessed by comparing their post-sample forecast performance with that of univariate ARMA-type models of marriage rates which are regarded as approximations to marriage rate final equation models. In this instance the ARM A models are preferred for forecasting purposes. The properties of the ARM A model forecasts are then examined and the modelling strategy is contrasted with the regression method used by Withers. |