Multi-attribute non-expected utility

In management applications of risk theory, planning and decision making are typically concerned with complex multi-dimensional attributes of risk and utility trade-offs between them. This paper presents a novel approach to multi-attribute non-expected utility which is especially designed to serve application and risk management purposes. It is based on a recently developed non-expected utility model that accommodates systematic violations of expected utility of various kinds observed in risky choice experiments. In the model, the possible outcomes of risky decisions are assumed to be multi-dimensional, that is, classified, measured, compared and assessed from different economic and non-economic perspectives simultaneously. Of the risk attributes to be jointly evaluated in a decision problem, each is supposed to be utility independent of the complementary set of all the other attributes also considered. Mutual utility independence and additive independence are particularly pronounced forms of utility independence. An order-preserving preference functional exists if the agent??s risk preferences satisfy familiar rationality requirements. The functional provides a consistently scaled, multi-linear representation in terms of single-attribute probability-dependent utility functions. Finally, the formalism is applied to explain observed trade-offs between monetary benefits obtained, and fatalities incurred, in the operation of large-scale industrial systems.     

Estimation of production risk and risk preference function: a nonparametric approach

While estimating parametric production models with risk, one faces two main problems. The first problem is associated with the choice of functional forms on the mean production function and the risk (variance) function. The second problem is associated with the specification of the risk preference function. In a parametric model the researcher chooses some ad hoc functional form on all these. It is obvious that the estimated (i) technology (mean production function), (ii) risk and (iii) risk preference functions are affected by the choice of functional form. In this paper we consider an estimation framework that avoids assuming parametric functions on all three. In particular, this paper deals with nonparametric estimation of the technology, risk and risk preferences of producers when they face uncertainty in production. Uncertainty is modeled in the context of production theory where producers’ maximize expected utility of anticipated profit. A multi-stage nonparametric estimation procedure is used to estimate the production function, the output risk function and the risk preference function. No distributional assumption is made on the random term representing production uncertainty. No functional form is assumed on the underlying utility function. Rice farming data from Philippines are used for an empirical application of the proposed model. Rice farmers are, in general, found to be risk averse; labor is risk decreasing while fertilizer, land and materials are risk increasing. The mean risk premium is about 3% of mean profit.     

基于效用函数的模糊数的比较与排序

对相同的模糊数进行比较,不同风险偏好的决策者,会得到不同的结论.效用函数是对风险偏好的度量,因此,模糊数的比较与排序的方法,一定要结合决策者的效用函数来构造.为此,根据效用函数定义了模糊效用函数,在此基础上定义了效用序.之后,证明效用序为全序,进一步利用结构元理论对效用序进行表述.根据效用函数反映风险偏好的程度,对效用序进行分类.这样,决策者对模糊数进行比较时,依据自身对风险偏好程度来选择效用序.     

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.     

How big is too big? Trading off the economies of scale of larger telecommunications network elements against the risk of larger outages

This paper builds a probabilistic model to analyze the risk–reward tradeoffs that larger telecommunications network elements present. Larger machines offer rewards in the form of cost savings due to economies of scale. But large machines are riskier because they affect more customers when they fail. Our model translates the risk of outages into dollar costs, which are random variables. This step enables us to combine the deployment cost and outage cost into a total cost. Once we express the decision makers’ preferences via a utility function, we can find the machine size that minimizes the total cost’s expected utility, thereby achieving an optimal tradeoff between reward and risk. The expected utility answers the question “how big is too big?”.     

An interactive decision procedure with multiple attributes under risk

Consider a finite set of alternatives under risk which have multiple attributes. MARPI is an interactive computer-based procedure to find an efficient choice in the sense of linear expected utility. The choice is based on incomplete information about the decision maker's preferences which is elicited and processed in a sequential way. The information includes qualitative properties of the multivariate utility function such as monotonicity, risk aversion, and separability. Further, in case of an additively separable utility function, bounds on the scaling constants are elicited, and preferences (not necessarily indifferences) between sure amounts and lotteries are asked from the decision maker. The lotteries are Bernoulli lotteries generated by MARPI using special strategies. At every stage of the procedure the efficient set of alternatives is determined with respect to the information elicited so far.The procedure has been fully implemented on a PC. The paper exhibits the basic ideas of MARPI and some details of its implementation.     

Scenario-based portfolio selection of investment projects with incomplete probability and utility information

In the selection of investment projects, it is important to account for exogenous uncertainties (such as macroeconomic developments) which may impact the performance of projects. These uncertainties can be addressed by examining how the projects perform across several scenarios; but it may be difficult to assign well-founded probabilities to such scenarios, or to characterize the decision makers’ risk preferences through a uniquely defined utility function. Motivated by these considerations, we develop a portfolio selection framework which (i) uses set inclusion to capture incomplete information about scenario probabilities and utility functions, (ii) identifies all the non-dominated project portfolios in view of this information, and (iii) offers decision support for rejection and selection of projects. The proposed framework enables interactive decision support processes where the implications of additional probability and utility information or further risk constraints are shown in terms of corresponding decision recommendations.     

A generalized decomposable multiattribute utility function

In this paper a generalized decomposable multiattribute utility function (MAUF) is developed. It is demonstrated that this new MAUF structure is more general than other well-known MAUF structures, such as additive, multiplicative, and multilinear. Therefore, it is more flexible and does not require that the decision maker be consistent with restrictive assumptions such as preferential independence conditions about his/her preferences. We demonstrate that this structure does not require any underlying assumption and hence solves the interdependence among attributes. Hence there is no need for verification of its structure. Several useful extensions and properties for this generalized decomposable MAUF are developed which simplify its structure or assessment. The concept of utility efficiency is developed to identify efficient alternatives when there exists partial information on the scaling constants of an assumed MAUF. It is assumed that the structure (decomposition) of the MAUF is known and the partial information about the scaling constants of the decision maker is in the form of bounds or constraints. For the generalized decomposable structure, linear programming is sufficient to solve all ensuing problems. Some examples are provided.     

Erratum: Betweenness Centralization for Bipartite Graphs

We address two related issues. First, we analyze the effects of risk preferences on cooperation in social dilemmas. Second, we compare social dilemmas in which outcomes represent gains with dilemmas where outcomes represent losses. We show that predictions on gain‐loss asymmetries with respect to conditions for cooperation crucially depend on assumptions concerning risk preferences. Under the assumption of risk aversion for gains as well as losses together with an assumption of decreasing absolute risk aversion, conditions for cooperation are less restrictive if outcomes represent losses than if outcomes represent gains. Conversely ‐ and counterintuitively ‐ under the assumption of S‐shaped utility, conditions for cooperation are more restrictive if outcomes represent losses than if outcomes represent gains. We provide an experimental test of such predictions. Only a minority of subjects behaves consistent with the assumption of S‐shaped utility. Furthermore, we find no empirical evidence for a general difference between cooperation in social dilemmas in which outcomes represent gains and dilemmas where outcomes represent losses. We do find evidence that risk preferences affect cooperation rates.     

The utility efficient set and its interactive reduction

We assume a decision situation under risk with incomplete information on preferences modelled as a vector utility function. We consider an additive aggregation of its components and partial information on the scaling constants. We develop the concept of utility efficiency to identify efficient strategies in discrete problems when the information about the scaling constants of the decision maker is in the form of a polyhedral cone. A characterization of the utility efficient set provides a practical way to compute such efficient strategies. We then discuss an interactive method based on the assessment of the scaling constants via an interactive paired comparison with its convergence. The method is complemented by a procedure to reduce the utility efficient set to aid in the process of reaching a final strategy.     

Interactive Utility Assessment in Multicriteria Decision Analysis Using Implicit Trade-off Information

This paper introduces a methodology for the assessment of a decision-maker's utility function, based on interactions requiring relatively easy responses of the implicit trade-off type, i.e. similar to responses required in STEM or goal programming methods. The estimation of the value-function model to represent preferences is useful in ranking or pruning elements of the decision space. Inputs required from the decision-maker are, however, less demanding than the pairwise comparisons (or similar preference statements) typically required by value-function models. The methodology thus appears to be appropriate for relatively large numbers of criteria. An algorithm for implementing the proposed methodology for finite action spaces is developed and applied to examples involving up to 15 criteria.     

基于决策者风险偏好大数据分析的大群体应急决策方法

针对重大突发事件应急决策大群体成员的风险偏好复杂难测问题,提出了一种新的基于决策者风险偏好大数据分析的大群体应急决策方法。首先专家群体对突发事件进行快速响应,生成若干应急预案及其风险属性信息;其次,社会公众通过网络等渠道参与到应急决策中来并形成决策大群体,给出不同预案的偏好值;然后,利用证据推理算法得出公众对各预案的风险效用值,将预案风险效用值与预案偏好值加权组合,得到各个预案的大群体决策者的风险偏好值;最后,基于风险偏好值,利用大数据分析技术对大群体的风险偏好进行聚类识别,从中筛选出风险中立者组成新的应急决策群体,再次聚类得出应急决策群体的成员组成结构,以此为基础计算决策者权重和应急预案的最终效用值,得应急预案排序结果。最后通过算例分析验证了方法的有效性和可行性。     

A Continuous-Risk Decision Analysis of Total Hip Replacement

Stochastic trees are stochastic processes, such as continuous-time Markov chains, which may be represented using tree diagrams. Continuous-risk utility assessment is a methodology that differs from standard assessment protocols by offering choices between alternatives which are not ‘sure’ things, thus enabling more realistic and familiar assessment scenarios to be posed. In this paper we develop, using information from the medical literature, stochastic tree models of the total hip replacement decision alternatives facing a patient with advanced osteoarthritis of the hip. We describe a decision analytic approach to the problem which uses continuous-risk utility assessment, and present empirical assessment results. We explore the sensitivity of the subjects' treatment recommendations to changes in model parameters and elicited preference values, offering insights into the variety of patient risk attitudes and preferences about functional well-being and the robustness of the total hip replacement decision to them.     

Multicriteria analysis: survey and new directions

Before the advent of multicriteria analysis, decision problems often took the form of the optimization of an objective function or a utility function. This approach has the advantage of resulting in well-defined mathematical problems but is not always representative of reality: in fact the comparison of several possible decisions is rarely made according to a single point of view and the preferences on a point of view are in many cases only modelled with difficulty by a function. The aim of this paper is to survey the different directions in which multicriteria analysis has developed: multiattribute utility theory, outranking relations, interactive methods. The methodology recently proposed by B. Roy is also presented.     

RISK-BASED MULTIATTRIBUTE DECISION-MAKING IN PROPERTY AND WATERSHED MANAGEMENT

ABSTRACT. Determining best management systems for properties and evaluating their sustainability at the watershed scale are useful and important aspects of integrated watershed management. Multiattribute decision-making (MADM) is very useful for modeling the selection of best management systems for properties in a watershed. This paper reviews four MADM approaches including utility theory, surrogate worth tradeoff, free iterative search and stochastic dominance with respect to a function (SDWF). Emphasis is on determining how the first three methods could be used to determine the best (most preferred) combinations of attributes and associated management systems for a property. An application of the expected utility method with risk neutral preferences is presented in which farmer's preferences for five attributes are used to rank five farming systems for an agricultural watershed in Missouri. A framework is presented for assessing the sustainability of the best management systems for all properties in a watershed and the cost-effectiveness of policies for enhancing sustainable resource management at the watershed scale.     

Moments of utility functions and their applications

This paper presents moments and cross-moments of utility functions and measures of utility dependence. We start with an interpretation of the nth moment of a utility function, and describe methods for its assessment in practice and consistency checks that need to be satisfied for any assessed moments. We then show how moments of a utility function (i) provide a new method to determine the parameters of a given functional form of a utility function and (ii) to derive the functional form of a utility function that satisfies some given moment assessments. Next, we derive a fundamental formula that relates the expected utility of a joint distribution to the expected utility of the marginal distributions for multiattribute utility functions. We use this formulation to provide an intuitive interpretation for cross-moments of utility functions and illustrate their use in (i) constructing multiattribute utility functions that incorporate utility dependence and (ii) in providing necessary conditions for utility independence in decisions with multiple attributes. We end with a new measure of utility dependence for multiattribute utility functions and work through several examples to illustrate the approach.     

Discounting axioms imply risk neutrality

Although most applications of discounting occur in risky settings, the best-known axiomatic justifications are deterministic. This paper provides an axiomatic rationale for discounting in a stochastic framework. Consider a representation of time and risk preferences with a binary relation on a real vector space of vector-valued discrete-time stochastic processes on a probability space. Four axioms imply that there are unique discount factors such that preferences among stochastic processes correspond to preferences among present value random vectors. The familiar axioms are weak ordering, continuity and nontriviality. The fourth axiom, decomposition, is non-standard and key. These axioms and the converse of decomposition are assumed in previous axiomatic justifications for discounting with nonlinear intraperiod utility functions in deterministic frameworks. Thus, the results here provide the weakest known sufficient conditions for discounting in deterministic or stochastic settings. In addition to the four axioms, if there exists a von Neumann-Morgenstern utility function corresponding to the binary relation, then that function is risk neutral (i.e., affine). In this sense, discounting axioms imply risk neutrality.     

Fuzzy preference orderings in group decision making

In this paper, some use of fuzzy preference orderings in group decision making is discussed. First, fuzzy preference orderings are defined as fuzzy binary relations satisfying reciprocity and max-min transitivity. Then, particularly in the case where individual preferences are represented by utility functions (utility values), group fuzzy preference orderings of which fuzziness is caused by differences or diversity of individual opinions are defined. Those orderings might be useful for proceeding the group decision making process smoothly, in the same manner as the extended contributive rule method.