A Novel Instantaneous Frequency Algorithm and Its Application in Stock Index Movement Prediction

Abstract

There are different approaches in defining instantaneous frequency in the literature; however, with a great variety of practical problems a specific approach is not always applicable. This paper proposes a new type instantaneous frequency suitable for a large class of signals called simple wave. The new instantaneous frequency is called -counting instantaneous frequency (IF) that should have independent interest. A stock index movement prediction approach is proposed in this paper based on the newly defined instantaneous frequency and signal decomposition. An experiment is provided based on the stock market data of Hong Kong Hang Seng Index. Effectiveness of the proposed approach is assessed through the comparison with BP neural network prediction method, and is proved to be promising.