Incorporating expert judgement into multivariate polynomial modeling Topic department: Decision support systems foundations

摘要

The study focuses on the conceptual approach to the systematic decision support in polynomial modeling of complex systems and deals with an unusual overlapping of mathematical modeling and decision theory. While incorporating non-formal expert judgement into empirical models, a whole new analytical framework emerged, which successfully resolved a number of new issues, such as formalization of expert knowledge, consistency of expert and empirical information, and generalization of the traditional concept of the “best” model. The described ideas are invariant over the empirical quality criteria and numerical methods used to estimate the coefficients of models. The purpose of the research was to combine expert knowledge and information extracted from the empirical data in order to construct models depicting the fundamental dynamics of the given system on a macro level. The paper provides a unified decision-making framework which can be used in decision support systems involving mathematical modeling.