Recently, increasing emphasis is being placed by government planners, corporate managers, military analyst and forecasters on the use of Scenario Analysis as a powerful tools aiding in decision making in the face of uncertainty. The idea behind this is to establish thinking about possible futures which can minimise surprises and broaden the span of managers' thinking about different possibilities, and when several interdependent events affect the future of an organisation, an industry, or a society, it is often useful to know how these events may affect each other. Determining the impact of external events on other such events, called a cross-impact analysis is usually accomplished by asking experts/knowledgeable people to discuss any relationship among the events and provide subjective estimates of conditional probabilities relating the events. However, there are possible problems, firstly, in some unfavourable economic and political environment because people may be reluctant to discuss the events openly and secondly, the subjective probability estimates may violate the law of probability theory, such as Bayes' theorem. Hence, this study has sought to investigate how the fusion and analysis of uncertainty or incomplete data through the use of Bayesian theory compares with people's intuitive judgements. These flexible, robust, and graphical probabilistic network are able to incorporate values from a wide range of sources including empirical values, experimental data and subjective values elicited from a number of experts. Bayesian-cross impact analysis provide a logical framework to combine each individual's set of one-at-a-time judgements, allowing comparisons with the same individuals' many-at-a-time direct intuitive judgements, this can be achieved through a series of fictitious and historical case studies. Building upon this, another area of interest would be the extent to which different elicitation can lead to equivalent or different judgements. In this study, competitive methods in Scenario Analysis will be used, these include; direct ranking of the variables perceived importance for discriminating between given hypotheses, likelihood ratios and conditional probabilities. Group Decision Support System (GDSS) is hereby presented to assist in eliciting anonymous comments and prepare a consistent probability estimates concerning interdependent events. Some general guidelines to consider when choosing, and using an appropriate scenario analysis method, including the merits and demerits of each approach will also be discussed. To cope with the issues of data shortage and linguistic expression of experts, the concept of fuzzy set theory will be considered. And hence, the GDSS developed through the use of subjective probability distribution will offer a flexible, robust methodology for the development of a normative model for the quantitative analysis of financial risk data.
Keywords: Scenario analysis; Conditional probability; Intuitive judgments; Linguistic expression; GDSS; Fuzzy Set Theory.
Biography: I am currently a PhD student at the University of Botswana