Pattern matching of time series and its application to trend prediction

Trend prediction of time series is an important research. Pattern matching provides a useful way for trend prediction. We mainly focus on the subsequence matching of time series in the paper. Firstly, we present the simulated series as the imputed pattern for the pattern matching; especially build a simulated ascending triangle series. Secondly, we propose an evaluation method with the actual trend of series to evaluate the experiment results. The proposed approach has been tested using a set of 1052 stocks, and the related assessments about the trend prediction are presented in the paper. The results show that the simulated series work better than the real series when used as the imputed pattern for trend prediction of time series.