Stock Price Prediction Using Kernel Adaptive Filtering Within a Stock Market Interdependence Approach

Stock prices are continuously generated by different data sources and depend on various factors such as financial policies and national economic growths. These financial time series are complex interconnected systems in which the price of one stock may be influenced by the economic factors of other stock markets. The prediction of stock prices, unlike traditional classification and regression problems, requires considering the sequential and interdependence nature of financial time series. This work proposes to sequentially predict stock prices using kernel adaptive filtering (KAF) within a stock market interdependence approach. Thus, unlike traditional approaches, stock prices are predicted using not only their local models but also the individual local models learned from other stocks, enhancing prediction performance. The proposed framework has been tested on 24 different stocks from three major economies. Simulation results show relatively low values of mean-square-error and better accuracy when compared with KAF-based methods