Publication Date: 2020/04/17
Abstract: The main objective of this research is to propose forecasting model of stock exchange (IDX) composite index using a weighted fuzzy time series (WFTS) model. The Mamdani inference system has been applied with the fuzzy model by using centroid defuzzification. After the models have been executed and verified, the performance of WFTS model has been compared with the conventional fuzzy time series (FTS) model using root mean square error (RMSE). The results showed that WFTS models had better performance than the conventional FTS models. The RMSE values achieved from WFTS and FTS models for training data sets were 0.314 and 0.4443, and for testing data 0.3246 and 0.4351, respectively. Finally, it is recommended that optimization techniques should be employed with the proposed type of models to improve their performance.
Keywords: Forecasting Model; Fuzzy Time Series; Weighted Fuzzy Time Series; Stock Exchange Composite.
DOI: No DOI Available
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT20MAR734.pdf
REFERENCES