Decision support system for portfolio management

Abstract

Decision making has remained a prime area of research, both for the management scientists and the behvioural scientists. Various approaches to decision making view either information, or analytical techniques, or intuition and judgment as vital to decision making, the segregation resulting from over—simplification. However, in practice, these segregated approaches have often been found to be inadequate for effective decision making, particularly, in solving complex, semi—structured decision problems. This led to the development of an integrated approach termed as decision support and the design of computer-based decision support system (DSS). The key mechanism used by a DSS in bringing about the integration is the dynamic interaction between the informational and computing power of the computers, the modeling capabilities of the analytical techniques, and the intuitive and judgmental abilities of human mind in a user—friendly, flexible environment. The objective of the present work was to explore the application of the concept of decision support to the complex and semi—structured decision problem of portfolio management in the Indian capital market environment taking advantage of the technological developments. Portfolio management is defined as a decision problem under uncertainty in the decision-theoretic terms, characterized by the lack of precise information about the outcomes and associated risk. The Indian capital market is characterized by a large number of small investors, sparingly available investment consultancy services, and a sudden burst of activity due to liberal government policies. It is imperative that a DSS for portfolio management in this context be developed around an inexpensive computer system which is easily accessible, and also that the software to be used be compatible and easily transportable to ensure wider use. Recent advances in the fields of computer science and information technology, particularly in terms of developments of inexpensive computing devices like personal computers, and non—procedural, user-friendly software fulfill the above requirements. The personal computer can be used as a vehicle for the spread of DSS, and the user—friendly software offer excellent design opportunities. This opened up not only portfolio management but a number of diverse fields all over the world for computer applications. The above considerations translated in terms of research problem meant the design and implementation of a DSS for portfolio management around a personal computer system using interactive, user-friendly software. IBM—compatible personal computer which has become an industry standard, and SYMPHONY — an integrated software with capabilities of spreadsheet, graphics, database, word-processing, and data communication have been used. The research methodology consists of 1. Evolving a methodology for DSS design which was envisaged to be application—specific and process-independent. Process—independence implies supporting a variety of decision processes. 2. Analyzing the analytical and behavioural aspects of portfolio management in Indian context which involved, apart from literature survey, a study of decision behaviour of several consultants and investors. 3. Designing a DSS and implementing it.

A DSS for portfolio management was designed to gain useful insights into the investors’ decision processes, and the design and implementation process of the DSS. The implication of the research findings include:

1. DSS-extends the bounds of rationality. 2. An integrated DIS design methodology tor application-specific, process-independent design has been evolved. 3. Personal computer is s viable DSS delivery vehicle. 4. The DSS supports a range of users, from laymen to experts. The experts can use the capabilities of the software to extend the scope of DSS. 5. The DSS can be used as a potential tool in education and research.

The limitations or the study as well as the directions for future research have also been discussed.