关键词: 模糊时间序列, 股票市场, 多步预测

Abstract: First, the universe of the discourse is divided into subintervals with the midpoints between two adjacent cluster centers generated by the fuzzy clustering method as their endpoints. And the sub-intervals are employed to fuzzify the time series into fuzzy time series. Then, the fuzzy time series model of high-order fuzzy relationships with multiple factors is built according to the main indexes representing stock price and trading volume. Finally, the model built in this paper is used to perform one-and multi-step forecasting of the daily Shanghai Stock Exchange composite index and Shenzhen Stock Exchange component index, respectively. Comparing with the benchmark model, one-step forecasting results show that the model improves the prediction accuracy and percent correct of the market up & down trend prediction, and multi-step forecasting results show that the model has good generalization.

Key words: fuzzy time series, stock market, multi-step forecast