An approach to mining financial markets through market state classification

Abstract

Financial markets have always been one of the most common application areas for a multitude of data mining techniques. Over the years a large number of autonomous prediction systems have been designed. In this paper an approach is proposed that offers a higher degree of control over the prediction process and over the exact market aspects that are being analyzed by the system. The concept can be applied to any trading strategy with the aim of enhancing forecasting accuracy. The paper also demonstrates how the theoretical concept can be applied in practice and concludes by illustrating the gains obtained by using the proposed approach.