A Novel Financial Decision Support Systems Based on Mendel Type-2 Fuzzy Set Theory

Abstract

Hedging against potential market risk and maximizing profit from stock trading are some of the driving forces for the development of financial decision support system.  Although presently many types of financial decision support available, there are still a lot of room for improvement. Hence, there is a significant importance to identify and develop a reliable financial decision support system with careful analysis on any critical issues. Accuracy has been the main goal of such decision support system, but it has been hampered by the biggest issue which is the uncertainty of the data. Fuzzy Neural Network (FNN) is a hybrid technique that combines fuzzy logic and neural network. FNN exploits the autonomy power of neural network, and therefore frees itself from the time-consuming process of knowledge base building. Despite all its advantages, FNN suffers from uncertainty when dealing with complex and highly uncertain event. The primary aim of this research work is to develop a financial decision support system based on type-2 fuzzy set. This is based on the knowledge that type-2 fuzzy set is a good alternative in reducing uncertainty in input data. The final application is an integrated FNN system, which has achieved the objective of reducing the general uncertainty of stock market model. Extensive experimental results based on several assumptions of approaches are presented for discussion.  Financial analysis foundations also briefly discussed in order to give a clearer prelude of the non-mathematical factor of this decision support system and its importance in being able to give a model which is considered to be the best input of the actual problem.