直觉模糊时间序列建模及应用

Modeling and application of IFTS

针对模糊时间序列对于预测不确定性的控制、有效的分区间隔和不同分区间隔达到一致的预测准确性方面研究的不足,构建了直觉模糊时间序列预测模型.新模型应用直觉模糊均值聚类算法优化序列区间划分,充分考虑数据点固有的模糊不确定性,较好地反映了系统的特征分布,提高了复杂环境中时间序列的预测性能且允许处理多因子预测问题.最后通过实例验证了所提出方法的有效性和优越性.

To the limitation of the research of controlling uncertainty in forecasting, effectively partitioning intervals and consistently achieving forecasting accuracy with different interval lengths, the intuitionistic fuzzy time series（IFTS） forecasting model is advanced. The novel forecasting model applies the intuitionistic fuzzy C＇-means（IFCM） clustering algorithm to optimize interval partitioning, which considers the data point fuzzy uncertainty fully, reflects the character distribution of uncertain system, enhances forecasting functionality in the complex environment better and allows the processing of multiple factors forecasting problems. Finally, a classical instance shows the effectiveness and superiority of the proposed method.