Information Fusion Technique for Weighted Time Series Model
Jia-Wen Wang, Ching‐Hsue Cheng · 2007
In this paper, we propose an information fusion technique for weighted time Series Model, is called OWA-MA forecasting model. The OWA-MA forecasting model combines OWA operator and weighted moving average (WMA). The model deals with the dynamical weighting problem more rationally and flexibly according to the situational parameter α value from the user's viewpoint. For verifying proposed method, we use two datasets to illustrate our performance, the datasets are: (1) dataset 1: the yearly data on enrollments at the university of Alabama [5] and (2) dataset 2: the forecast demand table to evaluate the proposed model [6]. Furthermore, the tracking signal as evaluation criteria to compares the proposed model with other models. It is shown that our proposed method proves better than other methods for time series model.