Abstract: | Summary This article describes the operational procedure for a mechanical monthly forecast of the money stock in The Netherlands for up to twelve months ahead. In addition to time series data of the money stock itself, the procedure uses the information provided by the disaggregation of the money stock into financial assets, sources of money supply and holdership. The forecast from a univariate ARIMA model of the money stock is combined with three different forecasts from vector ARIMA models for the components distinguished by the three ways of disaggregation. The combination weights, which differ for each number of months to be forecasted ahead, are determined by regression analysis.Professor of Economics at the Free University, Amsterdam and staff member of the Econometric Research and Special Studies Department of De Nederlandsche Bank NV, Amsterdam, respectively. This study was completed when the first author was at the Nederlandsche Bank. We gratefully acknowledge the useful comments of a referee on a previous version of this article. |