Real Time Estimation of Potential Output and Output Gap for the Euro-Area : Comparing Production Function with Unobserved Components and SVAR Approaches

Metadatas

Date

November, 2008

type
Language
Identifiers
License

info:eu-repo/semantics/OpenAccess


Keywords

potential output production function state-space models structural VARS C - Mathematical and Quantitative Methods/C.C3 - Multiple or Simultaneous Equation Models • Multiple Variables/C.C3.C32 - Time-Series Models • Dynamic Quantile Regressions • Dynamic Treatment Effect Models • Diffusion Processes • State Space Models E - Macroeconomics and Monetary Economics/E.E3 - Prices, Business Fluctuations, and Cycles/E.E3.E32 - Business Fluctuations • Cycles


Cite this document

Gian Luigi Mazzi et al., « Real Time Estimation of Potential Output and Output Gap for the Euro-Area : Comparing Production Function with Unobserved Components and SVAR Approaches », Hyper Article en Ligne - Sciences de l'Homme et de la Société, ID : 10670/1.fu3xfa


Metrics


Share / Export

Abstract 0

We develop a new version of the production function (PF) approach usually used for estimating the output gap of the euro area. Our version does not call for any (often imprecise) measure of the capital stock and improves the estimation of the trend total factor productivity. We asses this approach by comparing it with two other multivariate methods mostly used for output gap estimates, a multivariate unobserved components (MUC) model and a Structural Vector Auto-Regressive (SVAR) model. The comparison is conducted by relying on assessment criteria such as the concordance of the turning points chronology with a reference one, the inflation forecasting power and the real-time consistency of the estimates. Two contributions are achieved. Firstly, we take into account data revisions and their impact on the output gap estimates by using vintage datasets coming from the Euro Area Business Cycle (EABCN) Real-Time Data-Base (RTDB). Secondly, the PF approach, generally employed by policy-makers despite of its difficult implementation, is assessed. We thus improve on previous papers which limited their assessment on other multivariate methods, e.g. MUC or SVAR models. The different methods show different ranks in relation to the three criteria. This new PF estimate appears highly concordant with the reference chronology. Its forecasting power appears favourable only for the shortest horizon (1 month). Finally, the SVAR model appears more consistent in real-time.

From the same authors

On the same subjects

Similar documents

Within the same disciplines