Escudo de la República de Colombia
Sistema Nacional de Biliotecas - Repositorio Institucional Universidad Nacional de Colombia Biblioteca Digital - Repositorio Institucional UN Sistema Nacional de Bibliotecas UN

Nonlinear time series forecasting using mars

Velásquez Henao, Juan David and Franco Cardona, Carlos Jaime and Camacho, Paula Andrea (2014) Nonlinear time series forecasting using mars. DYNA; Vol. 81, núm. 184 (2014); 11-19 Dyna; Vol. 81, núm. 184 (2014); 11-19 2346-2183 0012-7353 .

Texto completo

[img]
Vista previa
PDF
432kB

URL oficial: http://revistas.unal.edu.co/index.php/dyna/article...

Resumen

One of the most important uses of artificial neural networks is to forecast non-linear time series, although model-building issues, such as input selection, model complexity and parameters estimation, remain without a satisfactory solution. More of research efforts are devoted to solve these issues. However, other models emerged from statistics would be more appropriated than neural networks for forecasting, in the sense that the process of model specification is based entirely on statistical criteria. Multivariate adaptive regression splines (MARS) is a statistical model commonly used for solving nonlinear regression problems, and it is possible to use it for forecasting time series. Nonetheless, there is a lack of studies comparing the results obtained using MARS and neural network models, with the aim of determinate which model is better. In this paper, we forecast four nonlinear time series using MARS and we compare the obtained results against the reported results in the technical literature when artificial neural networks and the ARIMA approach are used. The main finding in this research, it is that for all considered cases, the forecasts obtained with MARS are lower in accuracy in relation to the other approaches.

Tipo de documento:Artículo - Article
Información adicional:Derechos de autor reservados
Palabras clave:Artificial neural networks, comparative studies, ARIMA models, nonparametric methods.
Unidad administrativa:Revistas electrónicas UN > Dyna
Código ID:37737
Enviado por : Dirección Nacional de Bibliotecas STECNICO
Enviado el día :03 Julio 2014 16:43
Ultima modificación:04 Julio 2018 18:38
Ultima modificación:04 Julio 2018 18:38
Exportar:Clic aquí
Estadísticas:Clic aquí
Compartir:

Solamente administradores del repositorio: página de control del ítem

Vicerrectoría de Investigación: Número uno en investigación
Indexado por:
Indexado por Scholar Google WorldCat DRIVER Metabiblioteca OAIster BASE BDCOL Registry of Open Access Repositories SNAAC Red de repositorios latinoamericanos eprints Open archives La referencia Tesis latinoamericanas OpenDOAR CLACSO
Este sitio web se ve mejor en Firefox