Financial decision support with hybrid genetic and neural based modeling tools

This paper presents a comparative investigation of hybrid genetic classifiers vis-à-vis neural classifiers and statistical models in the financial domain. It is hypothesized that the proposed hybrid genetic classifier will perform better than the statistical counterpart. We provide a brief overview of the hybrid genetic classifier and discuss the design issues when applied to developing classification models for financial decision support. Further, the models are tested on a liquidation-merger problem. Results are consistent with the hypothesized premise. The proposed genetic classifiers outperform the statistical model. Implications of the comparison and issues for future research are addressed.