A decision theory for partially consonant belief functions

Partially consonant belief functions (pcb), studied by Walley, are the only class of Dempster-Shafer belief functions that are consistent with the likelihood principle of statistics. Structurally, the set of foci of a pcb is partitioned into non-overlapping groups and within each group, foci are nested. The pcb class includes both probability function and Zadeh’s possibility function as special cases. This paper studies decision making under uncertainty described by pcb. We prove a representation theorem for preference relation over pcb lotteries to satisfy an axiomatic system that is similar in spirit to von Neumann and Morgenstern’s axioms of the linear utility theory. The closed-form expression of utility of a pcb lottery is a combination of linear utility for probabilistic lottery and two-component (binary) utility for possibilistic lottery. In our model, the uncertainty information, risk attitude and ambiguity attitude are separately represented. A tractable technique to extract ambiguity attitude from a decision maker behavior is also discussed.     

On the confidence preferences model

In this paper we study the model of decision under uncertainty consistent with confidence preferences. In that model, a decision maker held beliefs represented by a fuzzy set of priors and tastes captured by a standard affine utility index on consequences. First, we find some interesting properties concerning the well-known maxmin expected utility model, taking into account the point of view of the confidence preferences model. Further, we provide new examples of preferences that capture ambiguity-averse attitudes weaker than ambiguity attitudes featured by maxmin expected utility theory. Finally, we discuss the axiomatic foundations for the confidence preferences model with optimistic behavior.     

Fuzzy optimality based decision making under imperfect information without utility

In the realm of decision making under uncertainty, the general approach is the use of the utility theories. The main disadvantage of this approach is that it is based on an evaluation of a vector-valued alternative by means of a scalar-valued quantity. This transformation is counterintuitive and leads to loss of information. The latter is related to restrictive assumptions on preferences underlying utility models like independence, completeness, transitivity etc. Relaxation of these assumptions results into more adequate but less tractable models. In contrast, humans conduct direct comparison of alternatives as vectors of attributes’ values and don’t use artificial scalar values. Although vector-valued utility function-based methods exist, a fundamental axiomatic theory is absent and the problem of a direct comparison of vectors remains a challenge with a wide scope of research and applications. In the realm of multicriteria decision making there exist approaches like TOPSIS and AHP to various extent utilizing components-wise comparison of vectors. Basic principle of such comparison is the Pareto optimality which is based on a counterintuitive assumption that all alternatives within a Pareto optimal set are considered equally optimal. The above mentioned mandates necessity to develop new decision approaches based on direct comparison of vector-valued alternatives. In this paper we suggest a fuzzy Pareto optimality (FPO) based approach to decision making with fuzzy probabilities representing linguistic decision-relevant information. We use FPO concept to differentiate “more optimal” solutions from “less optimal” solutions. This is intuitive, especially when dealing with imperfect information. An example is solved to show the validity of the suggested ideas.     

Generalizing variance to allow the inclusion of decision attitude in decision making under uncertainty

The problem of decision making under uncertainty is considered. It is noted that an alternative is described in terms of an uncertainty profile. We observe that a major difficulty in the decision process is the comparison of these uncertainty profiles. We discuss the need for introducing some features of an uncertainty profile to help simplify this comparison. We note that the quantification of these simplifying features involves some subjective considerations about the decision makers preferences. We introduce the idea of the decision maker’s attitudinal character to help in the formulation of these considerations. We then investigate two important features associated with an uncertainty profile. The first, the representative value, is a generalization of expected value commonly used under probabilistic uncertainty. The second, called the measure of deviation, provides a generalization of the concept of variance. We show how these new measures allows us to consider uncertainty profiles other then just the probabilistic one. They also allow us introduce other decision maker attitudes then the one implicitly assumed with the expected value and variance.     

A power law type of time preference on intertemporal choices

Discount utility, based on utility theory, is used to study human decision behaviors under the consideration of time preference. It assumes that by means of the axiomatic system of rationality, it is possible to quantify human beings’ utilities by some explicable models for intertemporal decision making. Recent studies have been based on two basic models: the exponential and the hyperbolic discount models. These two types of model have been proved to be either too fast for discounting or too restricted for fitting human beings’ discounting behaviors. In this study, a power law discount model is proposed. Axiomatic approach is used to ascertain the existence of the power law discount utility, and empirical investigations are implemented to verify the effectiveness of the proposed model.     

Decision making under uncertainty comprising complete ignorance and probability

This paper investigates a model of decision making under uncertainty comprising opposite epistemic states of complete ignorance and probability. In the first part, a new utility theory under complete ignorance is developed that combines Hurwicz–Arrow's theory of decision under ignorance with Anscombe–Aumann's idea of reversibility and monotonicity used to characterize subjective probability. The main result is a representation theorem for preference under ignorance by a particular one-parameter function – the τ-anchor utility function. In the second part, we study decision making under uncertainty comprising an ignorant variable and a probabilistic variable. We show that even if the variables are independent, they are not reversible in Anscombe–Aumann's sense. This insight leads to the development of a new proposal for decision under uncertainty represented by a preference relation that satisfies the weak order and monotonicity assumptions but rejects the reversibility assumption. A distinctive feature of the new proposal is that the certainty equivalent of a mapping from the state space of uncertain variables to the prize space depends on the order in which the variables are revealed. Explicit modeling of the order of variables explains some of the puzzles in multiple-prior model and the models for decision making with Dempster–Shafer belief function.     

A Structural Mapping Approach to Complex Decision-Making

Different methods currently available for multiple criteria decision analysis, such as cost-benefit analysis and utility theory, make strong axiomatic demands. The method suggested here uses multidimensional scaling techniques, as applied to the problem of constructing geographical maps from fragmentary information, to draw maps of policies involving many attributes in such a way as to throw most preferred and least preferred policies to opposite poles. The only axiomatic demand is non-transitive indifference. An analysis suggests that the method is robust against changes in the input data.     

On bargaining position descriptions in non-transferable utility games

We present a generalization to the Harsanyi solution for non-transferable utility (NTU) games based on non-symmetry among the players. Our notion of non-symmetry is presented by a configuration of weights which correspond to players' relative bargaining power in various coalitions. We show not only that our solution (i.e., the bargaining position solution) generalizes the Harsanyi solution, (and thus also the Shapley value), but also that almost all the non-symmetric generalizations of the Shapley value for transferable utility games known in the literature are in fact bargaining position solutions. We also show that the non-symmetric Nash solution for the bargaining problem is also a special case of our general solution. We use our general representation of non-symmetry to make a detailed comparison of all the recent extensions of the Shapley value using both a direct and an axiomatic approach.     

Towards a constructive approach to act-conditional subjective expected utility models

In structuring a decision problem under uncertainty, the uncertain environment may be affected by the choice of an act. In decision analysis, the decision maker provides subjective probabilities and utilities through separate elicitation processes, and then both components are combined together to give an index of his preference over decision alternatives. Based upon this conceptualisation of decision analysis, a constructive approach to act-conditional subjective expected utility theory is proposed. Two utility models have been addressed: the linear utility model and the weighted utility model.     

Nonprobabilistic entropies and indetermination measures in the setting of fuzzy sets theory

Our purpose in this paper is first of all to build an axiomatic generalization for the nonprobabilistic entropy of De Luca and Termini in the setting of fuzzy sets theory.We then build from this entropy an indetermination measure which can be used like discriminant function in Pattern Recognition when patterns are described by means of fuzzy sets.     

Using possibilistic logic for modeling qualitative decision: Answer Set Programming algorithms

A qualitative approach to decision making under uncertainty has been proposed in the setting of possibility theory, which is based on the assumption that levels of certainty and levels of priority (for expressing preferences) are commensurate. In this setting, pessimistic and optimistic decision criteria have been formally justified. This approach has been transposed into possibilistic logic in which the available knowledge is described by formulas which are more or less certainly true and the goals are described in a separate prioritized base. This paper adapts the possibilistic logic handling of qualitative decision making under uncertainty in the Answer Set Programming (ASP) setting. We show how weighted beliefs and prioritized preferences belonging to two separate knowledge bases can be handled in ASP by modeling qualitative decision making in terms of abductive logic programming where (uncertain) knowledge about the world and prioritized preferences are encoded as possibilistic definite logic programs and possibilistic literals respectively. We provide ASP-based and possibilistic ASP-based algorithms for calculating optimal decisions and utility values according to the possibilistic decision criteria. We describe a prototype implementing the algorithms proposed on top of different ASP solvers and we discuss the complexity of the different implementations.     

A Radon-Nikodym approach to measure information

We consider a decision maker facing uncertainty which behaves as a subjective expected utility maximizer. The value of information is traditionally captured as a greater expected utility the decision maker can achieve by selecting a best strategy as information arrives. We deal with the limit process of being better informed, and introduce an information density function depending solely on the states that gives an exact least upper bound to being more informed. This information density function is given by a Radon-Nikodym-type theorem for set functions and is explicitly computed for the countable case.     

Utility efficiency and its approximation

Summary We assume a multi-attribute decision making problem under uncertainty with partial information on the decision maker's preferences, by a vector utility function with two components and imprecision over their scaling constants. We propose an approximation set whose determination may be easier than the one of the utility efficient set and we consider an interactive procedure which uses such approximation to decision aid. We study some nesting and convergence properties based on the interactive reduction of the approximation set. Finally, we illustrate the procedure with a numerical example.     

Possibilistic sequential decision making

When the information about uncertainty cannot be quantified in a simple, probabilistic way, the topic of possibilistic decision theory is often a natural one to consider. The development of possibilistic decision theory has lead to the proposition a series of possibilistic criteria, namely: optimistic and pessimistic possibilistic qualitative criteria [7], possibilistic likely dominance [2], [9], binary possibilistic utility [11] and possibilistic Choquet integrals [24]. This paper focuses on sequential decision making in possibilistic decision trees. It proposes a theoretical study on the complexity of the problem of finding an optimal strategy depending on the monotonicity property of the optimization criteria – when the criterion is transitive, this property indeed allows a polytime solving of the problem by Dynamic Programming. We show that most possibilistic decision criteria, but possibilistic Choquet integrals, satisfy monotonicity and that the corresponding optimization problems can be solved in polynomial time by Dynamic Programming. Concerning the possibilistic likely dominance criteria which is quasi-transitive but not fully transitive, we propose an extended version of Dynamic Programming which remains polynomial in the size of the decision tree. We also show that for the particular case of possibilistic Choquet integrals, the problem of finding an optimal strategy is NP-hard. It can be solved by a Branch and Bound algorithm. Experiments show that even not necessarily optimal, the strategies built by Dynamic Programming are generally very good.     

Ranking residential properties by a multicriteria single price model

This study describes an application of the multicriteria single price model (Ballestero) to the ranking of alternatives. By a generalization of the original model, the equilibrium set of alternatives can be characterized from the viewpoints, respectively, of the demander and the supplier, and from that the efficiency index can be calculated. We demonstrate how, in a state of equilibrium, the two viewpoints result inevitably in inverse orders of ranking. In contrast with other proposals for full ranking of alternatives, the method used in the present study (i) assumes a moderate attitude on the part of the decision maker towards risk, with a robust axiomatic basis; (ii) assigns weights to the criteria independently of which alternative is being evaluated and the attitude (optimistic or pessimistic) of the decision maker; (iii) produces a cardinal hierarchy of the alternatives and not just an ordinal one. The model is illustrated by a sample of residential properties in the city of Valencia, Spain.     

Decisions under risk and uncertainty: A survey of recent developments

Summary This paper reviews recently proposed axiomatic models of choice under uncertainty and risk. The presentation focuses on the various models of transitive preferences which abandon the expected utility hypothesis by weakening the strong separability assumption known as the sure thing principle in the case of uncertainty and as the independence axiom in the case of risk. Special emphasis is on the remarkable similarities held in common by both the approaches to decisions under uncertainty and under risk.Paper presented at the International Conference on Operations Research, Vienna, Austria, August 28–31, 1990.     

Dynamic decision making without expected utility: An operational approach

Non-expected utility theories, such as rank dependent utility (RDU) theory, have been proposed as alternative models to EU theory in decision making under risk. These models do not share the separability property of expected utility theory. This implies that, in a decision tree, if the reduction of compound lotteries assumption is made (so that preferences at each decision node reduce to RDU preferences among lotteries) and that preferences at different decision nodes are identical (same utility function and same weighting function), then the preferences are not dynamically consistent; in particular, the sophisticated strategy, i.e., the strategy generated by a standard rolling back of the decision tree, is likely to be dominated w.r.t. stochastic dominance. Dynamic consistency of choices remains feasible, and the decision maker can avoid dominated choices, by adopting a non-consequentialist behavior, with his choices in a subtree possibly depending on what happens in the rest of the tree. We propose a procedure which: (i) although adopting a non-consequentialist behavior, involves a form of rolling back of the decision tree; (ii) selects a non-dominated strategy that realizes a compromise between the decision maker’s discordant goals at the different decision nodes. Relative to the computations involved in the standard expected utility evaluation of a decision problem, the main computational increase is due to the identification of non-dominated strategies by linear programming. A simulation, using the rank dependent utility criterion, confirms the computational tractability of the model.     

Decision analysis in clinical practice

Medical diagnosis and treatment, as examples of decision making under uncertainty, provide an ideal setting for the application of decision analysis. The paper, in selecting reports of medical decision analysis. discusses the usefulness of elements in decision analysis to the clinical setting. These include decision trees, probability assessments, and utility theory. Some concepts are welcomed by the medical profession and should be utilized to their utmost. However, some reported applications cannot be practically implemented for such reasons as physician time constraints and professional opposition. The paper identifies classes of clinical decision problems which are amenable to decision analysis and proposes ways of adapting the theory to clinical practice.