Time-Series Data Prediction Based on Trending Structure Sequence and Rough Set

Time series data is a series of observation data according to a certain time sequence. It has been penetrate various field. This paper applies rough set to the knowledge discovery of time series. The process of knowledge discovery in time series includes preprocessing of time series data, attributes selection and similarity sequence searching. Then, the time series is partitioned to a set of pattern (each pattern represents a trend of time series) by mobile window method. An information table is formed by the most important predicting attributes and target attribute which in the trending structure sequence identified from each pattern. This information table is suitable for the rough set to discover knowledge. The extracted rules can predict the time series behavior in the future. We demonstrate our method on time series stock market data.