Forecasting Gold Prices with Fuzzy Time Series.

The fuzzy time series was first introduced by Song and Chissom. Since then, it has become a more interesting subject among researchers. Fuzzy time series approaches are generally composing of three main stages such as fuzzification, determination of fuzzy relationships and defuzzification. Artificial neural networks (ANNs) have been successfully used in the fuzzy relationship determination stage of different fuzzy time series forecasting approaches. In the implementation, gold prices (USD/oz.) series is forecasted by fuzzy time series forecasting method based on particle swarm optimization. The results have shown that the predictive accuracy is improved. As a result of the application, it is seen that the method produces accurate forecasting results for the gold prices data. [ABSTRACT FROM AUTHOR]