A fuzzy time series model for Kuala Lumpur Composite Index forecasting

Study of fuzzy time series has increasingly attracted much attention due to its salient capabilities of tackling vague and incomplete data. A variety of forecasting models have devoted to improve forecasting accuracy. Recently, Chen, et al [12] has proposes a new model, which incorporates the concept of the Fibonacci sequence, the framework of Song and Chissom´s model and the weighted method of Yu´s model. However, the issue on lengths of intervals has not been investigated by Chen, et. al [12] despite Huarng [10] advocated that length of intervals affects forecasting results. Therefore the purpose of this paper is to propose a method of interval length into Chen, et. al [12] model and test it to Kuala Lumpur Composite Index (KLCI). Frequency density based partitioning of interval partitioning was incorporated into Chen et al model [12] and validate it using stock index KLCI data and check their performances. A two-year weekly period of KLCI stock index data was employed as experimental data sets. The results show that the frequency density based partitioning in the Chen et al model performed efficiently with the experimented datasets. This result reaffirms the importance of interval lengths in forecasting performances.