Fuzzy time series model for Egypt Gold Reserves forecasting based on fuzzy clustering

Fuzzy time series models have been used widely during the last decade. Such models have the ability to deal with historical data that are given in the form of linguistic values that can not be handled using traditional times series models. Fuzzy time series models have been used successfully in various problems such as: university enrolment, stock indices, and temperature prediction. In this paper, a fuzzy time series model based on fuzzy clustering is presented. This model is an enhanced version of original model proposed by the authors in [1], [2]. The newly proposed model has been used to forecast the Egypt Gold Reserves based on official data starting from the first quarter of 2002 up to the first quarter of 2010. The comparison result with other fuzzy time series models as well as the traditional ARIMA model shows that the proposed model provides higher accuracy and efficient performance.