Over the last decade, a considerable literature on biologically inspired algorithms (BIA) has emerged. These powerful algorithms can be used for prediction and classification, and have clear application for use in financial modelling and in the development of trading systems. Financial markets represent a complex, ever-changing, environment in which a population of investors are competing for profit. Biological entities have long inhabited such environments, and have competed for resources to ensure their survival. It is natural to turn to algorithms which are inspired by biological processes to tackle the task of survival in a financial jungle. The UCD Natural Computing Research and Applications Group (NCRA) has an active research agenda in the application of BIAs to financial markets. Current projects range from the development of adaptive real-time trading systems, to the development of new optimisation and classification tools for financial modelling.

We recently launched the SFI Strategic Research Cluster on Financial Mathematics and Computation which features three NCRA Principal Investigators (Brabazon, Edelman and O'Neill).

 Biologically Inspired Algorithms for Financial Modelling Our latest book (published January 2006 by Springer) describes and demonstrates the application of a range of biologically inspired algorithms (BIAs) for the purposes of financial modeling. Part I of the book provides a thorough guide to the various bioinspired methodologies - neural networks, evolutionary computing (particularly genetic algorithms and grammatical evolution), particle swarm and ant colony optimization, and immune systems. Part II brings the reader through the development of models of market trading systems, defining the criteria and explaining how indicators should be analysed. Finally, Part III illustrates the application of the bioinspired methodologies to a range of real-world financial case studies, including the development of trading systems in equity and foreign exchange markets, the prediction of corporate bond ratings, and the assessment of corporate solvency. This book is aimed at two audiences: those in the finance community who wish to learn about advances in biologically inspired computing and how these advances can be applied to financial modelling; and those in the computer science community who wish to gain insight into the domain of financial modelling and trading system design.

Sample Chapters:

• Chapter 1
• Chapter 8

Yield Curve Modelling with Genetic Programming

Zheng Yin is currently pursuing research in the application of Genetic Programming for the purposes of yield curve modelling.

Financial Modelling with Estimation of Distribution Algorithms

Kai Fan is currently pursuing research in the application of Estimation of Distribution Algorithms (EDAs) and other evolutionary algorithms for the pricing of financial derivatives.These methodologies hold the potential of constructing fast, adaptive, optimisation tools which would be of particular utility in high-frequency, dynamic, environments such as financial markets. One strand of this work has been the application of Quantum-inspired Evolutionary Algorithms (QIEA) for the calibration of option pricing models and the analysis of the implied volatility surface for stock index options. QIEAs derive metaphorical inspiration from a blending of evolutionary and quantum concepts. In the latter case,concepts such qubits and the superposition of states hold out the possibility of developing fast optimising algorithms which are capable of tracking optima in dynamic environments.

Financial Derivatives Modelling using Biologically Inspired Algorithms

This research project is motivated by the growth of complicated financial derivative products and the accompanying demand for more efficient and effective pricing and hedging tools and is being undertaken by Jing Dang. The central questions addressed in this project are the design of algorithms for the efficient pricing and hedging of financial derivatives. Biological creatures have long inhabited complex, dynamic, environments and their activities can serve as a plausible inspiration for the design of powerful optimizing algorithms. These algorithms have demonstrated significant potential in solving a range of hard, real-world, problems in finance and beyond. This research focusses on the use of metaphors drawn from evolution and social systems for the above financial applications. Some of this work to date has involved the design of algorithms whose workings are inspired by the foraging behaviours (method s for locating, handling and ingesting food) of biological creatures. In forgaging, animals seek `value for money', in other words they implicitly seek to maximise the energy obtained per unit of time (cost) spent foraging, in the face of constraint s presented by its own physiology (e.g., sensing and cognitive capabilities) and environment (e.g., density of prey, risks fro m predators, physical characteristics of the search area). Not alone do animals such as lions and tigers engage in foraging be haviors but, stylistically speaking, far smaller organisms also engage in foraging-type behaviors. Although we usually do not think of bacteria of being complex creatures, they are in fact capable of highly sophisticated behaviours including foraging f or nutrients. In this project, algorithms inspired by bacterial behaviors have been successfully used for option model calibra tion.