Understanding and Predicting Dynamical Behaviours in Financial Markets: Financial Application Research in CERCIA

In this paper, we shall very briefly present some of our research work in computational finance that has been carried out so far since the inception of CERCIA. Six research studies with different subjects are summarised here. The techniques that we employ in the studies vary from evolutionary computation approaches (e.g. genetic programming), signal processing techniques (e.g. the power spectrum, the wavelet analysis, the correlation) to fractal analysis methods (e.g. detrended fluctuation analysis, fractal geometry). The applied areas range from stock market predictions, stock picking, automatic trading strategies and financial market understanding, etc. Some of the studies have already been published, whilst some are still in the pipeline and others are just in their initial stages. Thus, the quality of work is varied.