关键词: 直觉模糊集, 广义模糊时间序列, 记分函数, 均方误差

Abstract: This paper points out the deficiency of the traditional generalized fuzzy time series model in the sample data fuzzification, and introduces the theory of intuitionistic fuzzy sets to the traditional model. First of all, we increase a hesitant degree factor in the membership and the membership functions in order to make the sample data intuitionistic fuzzy, and it reflects the uncertainty of the actual data more exquisite. Then, we use score function to describe the membership in a simplified model. In the framework of traditional generalized model, it builds a generalized fuzzy time series model based on intuitionistic fuzzy. At last, the typical Alabama university enrollment is chosen as the forecasting targets, and the empirical results show that the proposed model greatly outperforms the conventional counterparts.

Key words: intuitionistic fuzzy set, generalized fuzzy time series, score function, mean squared error