View article

This chapter presents a novel application of a neuro-evolutionary methodology for the purposes of intraday trading of German government bond futures. This market is very liquid, is the focus of substantial trading activity, and hence, quality trading systems for this market have considerable practical implication. To date, few studies have examined the potential utility of computational intelligence methodologies for the purpose of trading on this market and none have adopted a neuro-evolutionary approach. Our results suggest that structure in the time series of bond futures prices can be uncovered and that this information can be used to trade with some success. A number of future extensions of this study are also indicated.