![]() |
|||
![]()
|
![]() |
![]() |
![]() |
Part II
|
1. | Introduction | ||
2. | Background | ||
3. | Applications | ||
3.1. | Stock Options Pricing -- AESOP | ||
3.1.1. | The Symbolic Model | ||
3.1.2. | Knowledge Representation | ||
3.1.3. | Rules | ||
3.1.4. | System Validation | ||
3.1.5. | System Performance | ||
3.1.6. | Summary | ||
3.2. | Commercial Loan Analysis | ||
3.2.1. | D&B Expert System | ||
3.2.2. | CREDEX | ||
3.2.3. | COMPASS | ||
3.2.4. | CUBUS | ||
3.2.5. | PARMENIDE | ||
3.2.6. | KABAL | ||
3.2.7. | EVENT | ||
3.3. | Capital Investment | ||
3.3.1. | Management Advisor | ||
3.3.2. | AFFIN | ||
3.3.3. | Cash Flow Profiler | ||
3.3.4. | PROJECT | ||
3.4. | Predicting Stock Market Behavior | ||
3.5. | Neural Network Applications | ||
3.6. | Hybrid Systems | ||
4. | Research Issues | ||
4.1. | Statistical, Model Development, and Validation Issues | ||
4.2. | Cognitive Issues | ||
4.3. | User Acceptance | ||
5. | Future Trends | ||
References | |||
Suggested Neural Network and Finance Internet Address |
The world of finance includes the very challenging problem of relating past information to what the future will bring. Decisions in the financial markets of bonds, stocks, currency, commodities, real estate, and decisions relating to capital budgeting, financial distress, etc. provide a steady diet of complex prediction and classification tasks that require the experience and expertise of the decision-maker. Consistently making efficient and reliable decisions is not easily achieved. The size, complexity and, in some cases, less frequent repetition of these financial judgment tasks also hinders development of expertise by an individual. One important tool used in financial judgement tasks is expert systems -- computer-based systems intended to provide intelligent judgments similar to those of human experts.
Expert system business applications started to emerge in the late 1970s, with the major growth occurring in the mid-1980s. In their review of business applications of expert systems, Wong and Monaco (1995a, b) found 214 articles that they classified as describing business expert systems. Finance was identified as the second largest application area, after Production/Operations. Their search involved the ABI/INFORM database and, in the Business Periodical Index (BPI) for the period January 1980 through December 1992, 24 textbooks on expert systems and related topics, and five journals that were not included in the ABI/INFORM database or the BPI, but were known to publish expert system articles.
In another review, Hayes-Roth and Jacobstein (1994) found that the area of Finance dominated the 87 systems described in the first four volumes of the Innovative Applications of Artificial Intelligence series during the period 1990 to 1992.
Eom (1996) compiled a bibliography of 291 papers that describes 440 operational ES applications. Similar to Wong and Monaco, he found that the largest number of operational expert systems are in production and operations management (47%), followed by finance (17%). Eom applied the criteria that the expert system be in actual use. He excluded prototype systems, systems being tested, expert systems that have fallen into disuse, and systems in non-business areas such as chemistry, physics, government, and medicine.
Coakes and Merchant (1996) surveyed 1000 companies across the U.K. in the Spring of 1994 to find out the use of information systems in general and expert systems in particular. A response rate of 22.5% showed that 50 out of 214 organizations use expert systems for a variety of applications, ranging from strategic to operational. About 24% of the respondents that claimed to use expert systems were in the financial service sector. The financial services companies used their systems primarily for routine activities and decision support. Interestingly, the study showed that it is not only large organizations that use expert systems -- 28% of the respondents using them had less than 200 employees.
This chapter describes a sample of finance expert systems that are either in use or have given very promising results in the prototype stage and can therefore be predicted to proceed to implementation. The specific aspects of finance covered in the chapter are stock options pricing, commercial loan analysis (or credit granting or financial risk assessment), capital budgeting (investments), and stock market prediction, along with finance-related applications of neural networks-based systems.
Previous | Table of Contents | Next |
![]() |
|
Use of this site is subject certain Terms & Conditions. Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited. Please read our privacy policy for details. |