Application based Time Series Prediction using SQP Optimized Fuzzy Systems

This paper presents a fuzzy approach to the prediction of highly nonlinear time-series. The optimized Mamdani-type fuzzy system is applied for the input-output modelling of measured data. In order to tune fuzzy membership functions, a sequential quadratic programming (SQP) method is employed. The proposed method is evaluated and validated on a highly complex time series, daily gold price data. Month selection in each stage is based on computed correlation coefficients. The simulation results for one-step-ahead prediction of daily gold price in 2010 demonstrate better performance of the applied structure until the higher significant dynamics of the chaotic trend is taken into account