AN INTELLIGENT ASSISTANT FOR FINANCIAL HEDGING

Abstract

Problems in Finance, particularly those involving risk assessment and management, have been slow to yield to expert systems technology for two reasons. First, expert reasoning in such problems is often based on “first principles" instead of “situation-action" rules that characterize most expert systems. Secondly, the knowledge involved, such as that about financial instruments, is constantly changing. This would make it extremely difficult to keep a rule-base accurate. We have developed a representation in the domain of financial hedging that has the following characteristics. First, it allows for reasoning qualitatively based on first principles using the fundamental quantitative valuation models that characterize each instrument. Secondly, it uses object oriented concepts and inheritance to minimize the effort needed to set up the knowledge base and keep it current. Thirdly, it includes a calculus for derivation of qualitative knowledge of "one-dimensional-order", which allows it to solve problems where optimality constraints are qualitative.