Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/153797
Authors: 
González, Andrés
Hubrich, Kirstin
Teräsvirta, Timo
Year of Publication: 
2011
Series/Report no.: 
ECB Working Paper 1363
Abstract: 
We propose a new method for medium-term forecasting using exogenous information. We first show how a shifting-mean autoregressive model can be used to describe characteristic features in inflation series. This implies that we decompose the inflation process into a slowly moving nonstationary component and dynamic short-run fluctuations around it. An important feature of our model is that it provides a way of combining the information in the sample and exogenous information about the quantity to be forecast. This makes it possible to form a single model-based inflation forecast that also incorporates the exogenous information. We demonstrate, both theoretically and by simulations, how this is done by using the penalised likelihood for estimating the model parameters. In forecasting inflation, the central bank inflation target, if it exists, is a natural example of such exogenous information. We illustrate the application of our method by an out-of-sample forecasting experiment for euro area and UK inflation. We find that for euro area inflation taking the exogenous information into account improves the forecasting accuracy compared to that of a number of relevant benchmark models but this is not so for the UK. Explanations to these outcomes are discussed.
Subjects: 
Nonlinear forecast
nonlinear model
nonlinear trend
penalised likelihood
structural shift
time-varying parameter
JEL: 
C22
C52
C53
E31
E47
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.