A comparison of exponential smoothing and neural networks in time series prediction

In this article, we compare the accuracy of the forecasts for the exponential smoothing (ES) approach and the radial basis function neural networks (RBFNN) when three nonlinear time series with trend and seasonal cycle are forecasted. In addition, we consider the recommendations of preprocessing by...

Full description

Autores:
Velásquez Henao, Juan David
Zambrano Pérez, Cristian Olmedo
Franco Cardona, Carlos Jaime
Tipo de recurso:
Article of journal
Fecha de publicación:
2013
Institución:
Universidad Nacional de Colombia
Repositorio:
Universidad Nacional de Colombia
Idioma:
spa
OAI Identifier:
oai:repositorio.unal.edu.co:unal/74156
Acceso en línea:
https://repositorio.unal.edu.co/handle/unal/74156
http://bdigital.unal.edu.co/38633/
Palabra clave:
Forecasts combination
nonlinear models
artifi cial neural networks
nonlinear time series
Rights
openAccess
License
Atribución-NoComercial 4.0 Internacional
Description
Summary:In this article, we compare the accuracy of the forecasts for the exponential smoothing (ES) approach and the radial basis function neural networks (RBFNN) when three nonlinear time series with trend and seasonal cycle are forecasted. In addition, we consider the recommendations of preprocessing by eliminating the trend and seasonal cycle using simple and seasonal differentiation. Finally, we use forecast combining for determining if there is complementary information between the forecasts of the individual models. Our numerical evidence supports the following conclusions: ES models have a better fi t but lower predictive power than the RBFNN; detrending and deseasonality allows the RBFNN to fi t and forecast with more accuracy than the RBFNN trained with the original dataset; there is no evidence of information complementarity in the forecasts such that the methodology of forecasts combination is not able to predict with more accuracy than the RBFNN and ES methodologies.