V. Brunhes-Lesage

Nowcasting German GDP: A comparison of bridge and factor models

Governments and central banks need to have an accurate and timely assessment of Gross Domestic Products (GDP) growth rate for the current quarter, as this is essential for providing a reliable and early analysis of the current economic situation. This paper presents a series of models conceived to forecast the current German GDPs quarterly growth rate. These models are designed to be used on a monthly basis by integrating monthly economic information through bridge models, thus allowing for the economic interpretation of the data. We do also forecast German GDP by dynamic factor models. The combination of these two approaches allows selecting economically relevant explanatory variables among a large data set of hard and soft data. In addition, a rolling forecast study is carried out to assess the forecasting performance of the estimated models. To this end, publication lags are taken into account in order to run pseudo out-of-sample forecasts. We show that it is possible to get reasonably good estimates of current quarterly GDP growth in anticipation of the official release, especially from bridge models.

Monthly Forecasting of French GDP: A Revised Version of the Optim Model

This paper presents a revised version of the model OPTIM, proposed by Irac and Sedillot (2002), used at the Banque de France in order to predict French GDP quarterly growth rate, for the current and next quarters. The model is designed to be used on a monthly basis by integrating monthly economic information through bridge models, for both supply and demand sides of GDP. For each GDP component, bridge equations are specified by using a general-to-specific approach implemented in an automated way by Hoover and Perez (1999) and improved by Krolzig and Hendry (2001). This approach allows to select explanatory variables among a large data set of hard and soft data. The final choice of equations relies on a recursive forecast study, which also helps to assess the forecasting performance of the revised OPTIM model in the prediction of aggregated GDP. This study is based on pseudo real-time forecasts taking publication lags into account. It turns out that the model outperforms benchmark models.

New Estimate of the MIBA Forecasting Model. Modeling First-Release GDP Using the Banque De France's Monthly Business Survey and the 'Blocking' Approach

This paper introduces the new Monthly Index of Business Activity (MIBA) model of the Banque de France for forecasting Frances GDP. As the previous versions, the model relies exclusively on data from the monthly business survey (EMC) conducted by the Banque de France. However, several major changes have been implemented in the present version, such as the shift from a model based on factors to a model based on survey opinions, the explicit targeting of first-release GDP, and the use of the “blocking” approach to deal with mixed frequencies and missing observations. The selected monthly equations are consistent with the time frame of real-time forecasting exercises: the first month equation is dominated by data on expected evolution of the economic activity across the coincident quarter, while for the second and third month equations data on observed economic activity become more important and forward-looking information is progressively discarded. Finally, out-of-sample results suggest that the new MIBA model broadly outperforms several competing models, such as the previous version of MIBA and models based on alternative specifications.

The Monthly Index of Business Activity (MIBA): A Revision

This paper proposes new bridge equations for the Monthly Index of Business Activity (MIBA) published by the Banque de France. The MIBA is a forecasting tool for the quarterly GDP growth in France both for the current quarter and the next quarter, originally based on the surveys in the industrial sector published in the Monthly Business Survey (MBS) conducted by the Banque de France. Two improvements are suggested: first, from a technical viewpoint, we use an automatic model selection procedure which brings a robust, clear and systematic framework for selecting variables; second, from a modelling viewpoint, we take into account the business surveys in the services sector published by the Banque de France. The forecasting performance of the different models is evaluated.