A Multiobjective Expert System for Suppliers of Out-of-the-Money Options
Loading...
Links to Files
Permanent Link
Collections
Author/Creator
Author/Creator ORCID
Date
1984
Type of Work
Department
Program
Citation of Original Publication
Morse J.N. (1984) A Multiobjective Expert System for Suppliers of Out-of-the-Money Options. In: Grauer M., Wierzbicki A.P. (eds) Interactive Decision Analysis. Lecture Notes in Economics and Mathematical Systems, vol 229. Springer, Berlin, Heidelberg
Rights
Subjects
Abstract
The proliferation of telecommunications and computing capability has sparked speculation about dispersing work. The electronic cottage industry approach has potential in the area of office and professional services. This paper proposes that certain functions of a centralized financial market can be decentralized. To relocate an element of an information processing system, its functions as well as its linkages must be duplicated. For that purpose, we draw on the theories of multiple criteria decision making (MCDM), artificial intelligence, and finance. On the trading floors of the world’s options exchanges, people called market-makers provide liquidity; when a buyer cannot find a seller, or a seller cannot find a buyer, these functionaries sometimes fill that void by trading for their own account. The physically grueling work of transacting on the exchange floor is often delegated to lower level employees. These people act with supervision from senior people, or perhaps with a set of decision rules formulated by the suppliers of capital. This paper will attempt to outline the features of an automated, remotely-sited system that emulates the market-making function which is normally performed on the site of centralized options exchanges. To make the task less formidable, I will concentrate on one very specialized market-making function, namely supplying uncovered out-of-the money options. The system is intended to promote useful communication between man and computer. Although it would be presumptuous to say that the system can learn, it can decide to query a human when certain conditions are present. From these situations, a richer array of decision rules will develop.