Home > Research > Publications & Outputs > A new approach based on artificial neural netwo...
View graph of relations

A new approach based on artificial neural networks for high order bivariate fuzzy time series

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Publication date1/01/2009
Host publicationApplications of Soft Computing: From Theory to Praxis
Number of pages9
ISBN (print)9783540896180
<mark>Original language</mark>English

Publication series

NameAdvances in Intelligent and Soft Computing
ISSN (Print)1867-5662


When observations of time series are defined linguistically or do not follow the assumptions required for time series theory, the classical methods of time series analysis do not cope with fuzzy numbers and assumption violations. Therefore, forecasts are not reliable. [8], [9] gave a definition of fuzzy time series which have fuzzy observations and proposed a forecast method for it. In recent years, many researches about univariate fuzzy time series have been conducted. In [6], [5], [7], [4] and [10] bivariate fuzzy time series approaches have been proposed. In this study, a new method for high order bivariate fuzzy time series in which fuzzy relationships are determined by artificial neural networks (ANN) is proposed and the real data application of the proposed method is presented.