抽象的な

Type-2 fuzzy systems theory and its application in the stock trend analysis

Zhao Juan, Jiang Lihui

Fuzzy time series model is to solve the classical time series analysis methods can not deal with the fuzzy problem and the birth of. Along with the people to solve complex problems to be increasing, the research is more and more thorough, its application is more and more widely. From the theoretical model, establishing fuzzy relationship matrix, extraction and prediction rules and improve forecasting accuracy has been the focus of the study of fuzzy time series. A fuzzy time series model, C- AFS theory of fuzzy decision tree and based on evidence theory. Based on fuzzy decision tree and evidence theory in the introduction of AFS theory, C-, fuzzy time series model to determine the low prediction rules in the fuzzy relation, fuzzy time series combined with these three methods of forecasting model. The experimental results show that the generalized fuzzy time series model can best sum up the fuzzy sequence contains information which get the forecasting result comparison of similar model better