Decision theory and rules of thumb (2014)

Abstract

This chapter presents a relatively new and rapidly developing interdis- ciplinary theory of decision making, the theory of fast and frugal heuristics. It is first shown how the theory complements most of the standard theories of decision making in the social sciences such as Bayesian expected utility theory and its vari- ants: Fast and frugal heuristics are not derived from normatively compelling axioms but are inspired by the simple rules of thumb that people and animals have been empirically found to use. The theory is illustrated by presenting the basic concepts and mathematics of some fast and frugal heuristics such as the recognition heuristic, the take-the-best heuristic, and fast and frugal trees. Then, applications of fast and frugal heuristics in a number of problems are described (how do laypeople make investment decisions? how do military staff identify unexploded ordnance buried in the ground? how do doctors decide whether to admit a patient to the emergency care or not?) It is emphasized that there are no good or bad decision models per se but that all models can work well in some situations and not in others, and thus the goal is to find the right model for each situation. Accordingly, in all applications, the performance of fast and frugal heuristics is compared, by computer simulations and mathematical analyses, to the performance of standardmodels such as Bayesian networks, classification-and-regression trees and support-vector machines. Finally, ways of combining standard decision theory and rules of thumb are discussed

Bibliographic entry

Katsikopoulos, K. V. (2014). Decision theory and rules of thumb. In W. Pedrycz & P. Guo (Eds.), Human-centric decision-making models for social sciences (Studies in Computational Intelligence No. 502) (pp. 75-96). Heidelberg: Springer.

Miscellaneous

Publication year 2014
Document type: In book
Publication status: Published
External URL:
Categories:
Keywords: mathematical modeling

Edit | Publications overview