Department of Economics and Business Economics

Timo Teräsvirta

Forecasting macroeconomic variables using neural network models and three automated model selection techniques

Research output: Research - peer-reviewJournal article

Standard

Forecasting macroeconomic variables using neural network models and three automated model selection techniques. / Kock, Anders Bredahl; Teräsvirta, Timo.

In: Econometric Reviews, Vol. 35, No. 8-10, 2016, p. 1753-1779.

Research output: Research - peer-reviewJournal article

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{94404bd7e5224f44b380d06e6a5432da,
title = "Forecasting macroeconomic variables using neural network models and three automated model selection techniques",
abstract = "When forecasting with neural network models one faces several problems, all of which influence the accuracy of the forecasts. First, neural networks are often hard to estimate due to their highly nonlinear structure. To alleviate the problem, White (2006) presented a solution (QuickNet) that converts the specification and nonlinear estimation problem into a linear model selection and estimation problem. We shall compare its performance to that of two other procedures building on the linearization idea: the Marginal Bridge Estimator and Autometrics. Second, one must decide whether forecasting should be carried out recursively or directly. This choice is investigated in this work. The economic time series used in this study are the consumer price indices for the G7 and the Scandinavian countries. In addition, a number of simulations are carried out and results reported in the article.",
keywords = "Artificial neural network, Forecast comparison, Model selection, Nonlinear autoregressive model, Nonlinear time series, Root mean Square forecast error, Wilcoxon's signed-rank test",
author = "Kock, {Anders Bredahl} and Timo Teräsvirta",
year = "2016",
doi = "10.1080/07474938.2015.1035163",
volume = "35",
pages = "1753--1779",
journal = "Econometric Reviews",
issn = "0747-4938",
publisher = "Taylor & Francis Inc.",
number = "8-10",

}

RIS

TY - JOUR

T1 - Forecasting macroeconomic variables using neural network models and three automated model selection techniques

AU - Kock,Anders Bredahl

AU - Teräsvirta,Timo

PY - 2016

Y1 - 2016

N2 - When forecasting with neural network models one faces several problems, all of which influence the accuracy of the forecasts. First, neural networks are often hard to estimate due to their highly nonlinear structure. To alleviate the problem, White (2006) presented a solution (QuickNet) that converts the specification and nonlinear estimation problem into a linear model selection and estimation problem. We shall compare its performance to that of two other procedures building on the linearization idea: the Marginal Bridge Estimator and Autometrics. Second, one must decide whether forecasting should be carried out recursively or directly. This choice is investigated in this work. The economic time series used in this study are the consumer price indices for the G7 and the Scandinavian countries. In addition, a number of simulations are carried out and results reported in the article.

AB - When forecasting with neural network models one faces several problems, all of which influence the accuracy of the forecasts. First, neural networks are often hard to estimate due to their highly nonlinear structure. To alleviate the problem, White (2006) presented a solution (QuickNet) that converts the specification and nonlinear estimation problem into a linear model selection and estimation problem. We shall compare its performance to that of two other procedures building on the linearization idea: the Marginal Bridge Estimator and Autometrics. Second, one must decide whether forecasting should be carried out recursively or directly. This choice is investigated in this work. The economic time series used in this study are the consumer price indices for the G7 and the Scandinavian countries. In addition, a number of simulations are carried out and results reported in the article.

KW - Artificial neural network

KW - Forecast comparison

KW - Model selection

KW - Nonlinear autoregressive model

KW - Nonlinear time series

KW - Root mean Square forecast error

KW - Wilcoxon's signed-rank test

U2 - 10.1080/07474938.2015.1035163

DO - 10.1080/07474938.2015.1035163

M3 - Journal article

VL - 35

SP - 1753

EP - 1779

JO - Econometric Reviews

T2 - Econometric Reviews

JF - Econometric Reviews

SN - 0747-4938

IS - 8-10

ER -