Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/175972 
Year of Publication: 
2004
Series/Report no.: 
Texto para discussão No. 485
Publisher: 
Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio), Departamento de Economia, Rio de Janeiro
Abstract: 
In this paper we examine the forecast accuracy of linear autoregressive, smooth transition autoregressive (STAR), and neural network (NN) time series models for 47 monthly macroeconomic variables of the G7 economies. Unlike previous studies that typically consider multiple but fixed model specifications, we use a single but dynamic specification for each model class. The point forecast results indicate that the STAR model generally outperforms linear autoregressive models. It also improves upon several fixed STAR models, demonstrating that careful specification of nonlinear time series models is of crucial importance. The results for neural network models are mixed in the sense that at long forecast horizons, an NN model obtained using Bayesian regularization produces more accurate forecasts than a corresponding model specified using the specific-to-general approach. Reasons for this outcome are discussed.
Subjects: 
forecast combination
forecast evaluation
neural network model
nonlinear modelling
nonlinear forecasting JEL Codes: C22
C53
Document Type: 
Working Paper

Files in This Item:
File
Size
357.24 kB





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