Back
Forecasting models currently applied to indicators computed on the basis of surveys results
European Commission
2002 / 2003
Principal investigator
Jordi Suriñach
Investigative team
Jordi Suriñach / Manuel Artís / Miquel Clar / Òscar Clavería / Juan Carlos Duque / Rosina Moreno / Ernest Pons / Raül Ramos
Objective
Comparing different time series methods for short-run forecasting of Business and Consumer Surveys Indicators for the Euro area. In this context, the effects of the seasonal adjustment procedure of original data were also analysed. Analysing the possibility of forecasting some of the main quantitative macroeconomic variables of the Euro area using the information given by the Business and Consumer Surveys Indicators. Providing a methodological guide for the monthly use of the models, the calculation of forecasts and the preparation of the basic results for a report on the outlook of the Euro area economy.
Methodology
Forecasting Business and Consumer Surveys Indicators: Assessing the forecasting accuracy of different methods and models. The effects of seasonality on data revision and on forecast accuracy. The effects of removing outliers using Tramo/Seats on data revision and on forecast accuracy. Forecasting quantitative variables using information from Business and Consumer Surveys Indicators: Benchmark models. “Augmented” autoregression, Markov Switching Regime and VAR models. Leading indicators models and quantification of expectations.
Reference:
Tender ECFIN/2002/A3-01. Contract number: Contract B2002/A3500/57 Theme:
Macroeconomic simulations and forecasts Ambit:
European