Decision Making Processes Under Uncertainty: An Econometric Analysis

Expected Utility and Sequential Elimination Models of Career Decision Making

Muth’s Rational Expectations Hypothesis, That “…The Subjective Probability Distributions Are Distributed Around an Objective Probability Distribution, for a given Information Set…” is an Oxymoron. It Can Be Corrected by Amending It to Read “…The Subjective Probability Distributions Are Distributed Around a Subjective Consensus Probability Distribution, for a given Information Set…”

One important aspect of risk analysis is the individual and societal comparison of probability distributions over a set of undesirable outcomes. To compare all possible distributions it is mandatory to assign one single value (called an index) to each distribution. Typical indices are the expected value and the expected utility model; the paper notes some theoretical and practical limitations of these indices. The paper briefly describes the author’s linearized moments model (LMM) that gives consideration to individual risk perceptions and that takes into account the shape of the probability distribution including the low-probability high-consequence tail. The LMM solves the main theoretical and practical inconsistencies of the expected utility model. It is noted that current indices like the expected value, the expected utility model and the prospect theory are special cases of the LMM.

This paper surveys nearly two hundred scholarly works that use mathematical methods, which include stochastic models, difference and differential equation models, expected utility models, and various types of game theoretic models, to study domestic political conflict (DPC), which includes terrorism, guerrilla wars and insurrections. A citation count reveals that the DPC articles surveyed here cite less than three quarters of an article from within their own DPC modelling tradition and cite less than two articles from any DPC modelling tradition. The only exceptions to the rule that “nobody cites nobody else” are the stochastic and expected utility modelers. I conclude that the “field” of formal models of DPC hardly exists: few authors read other authors, few articles cite other articles, few models build on other models. Several suggestions aimed at promoting greater accumulation in formal models of DPC are offered.

We analyze three explanations of candidate choice in presidential nomination campaigns: (1) a preference model, which contends that nomination choice is based on the ideological, issue, or candidate preferences of the voter; (2) a candidate-chances explanation that argues choice is motivated by the chances candidates have of winning the nomination or the general election; and (3) an expected utility model that includes both preferences and candidate chances by discounting ideological, issue, and candidate-trait preferences by the perceived chances each candidate has of winning in November. We test these explanations on samples of Democratic caucus attenders and state convention delegates from Iowa in 1984. The expected utility model most successfully predicts candidate choice in both samples. Even with appropriate controls for candidate affect, both preference and candidate chances measures have significant independent effects on candidate choice. Our analysis suggests that the interests of individual nomination participants and party organizations may not be as disparate as many critics of the contemporary process believe.

The opportunity cost of mean–variance choice under estimation risk

The dominant theory of individual decision making under uncertainty is the von Neumann and Morgenstern's expected utility model (EUM). Although, this model has gained its popularity from its ability to explain a wide range of attitudes toward risk, empirical evidence over the last three decades has documented paradoxical behavior that is inconsistent with the EUM model. The present study examines in an experimental setting individuals' inconsistent behavior in choosing between two pairs of lotteries known as the Allais paradox. This study extends previous literature on money gamble situations by conducting a series of experiments to test expected utility behavior on auditing student subjects in pure money and audit settings. Two variants of audit settings are examined: overall audit plan case and audit procedure evaluation of internal control case.

Résumé G. Becker avance que le modèle de l’utilité espérée (MUE) est le plus efficace que les sciences sociales puissent utiliser. La raison principale de son importance est qu’il engendre des explications autosuffisantes. Mais il est impuissant à expliquer de nombreux phénomènes sociaux. La raison en est que les actions s’appuient couramment sur des croyances non triviales dont il faut rendre compte dès lors qu’on vise une explication autosuffisante. Or on croit couramment que faire quelque chose est bon, juste, vrai, légitime, etc. sur la base de raisons non instrumentales. La rationalité ne doit donc pas être conçue comme étant de caractère exclusivement instrumental. Smith et Weber parmi d’autres ont repéré ce point et esquissé un modèle qui repose sur une définition générale de la rationalité (MRG) et conserve la propriété d’autosuffisance du MUE tout en éliminant les sources de ses limitations. Plusieurs exemples d’application de ce modèle sont présentés.

Rational choice theory (RCT), an influential theory in the study of American politics and a well-established theory in the field of international relations, only gained sustained attention within comparative politics beginning in the early 1990s. This chapter presents a definition of RCT. It suggests that students of comparative politics should strive to assimilate the emphasis on actors and instrumental rationality that are a characteristic of RCT, but go beyond RCT and built a broader, encompassing theory of action. The chapter shows that the core theoretical principles of game theory are limited, both in terms of the domains where it is applicable and its explanatory power in the domains where it is applicable. Thus, in an effort to assess the strengths and weaknesses of game theory, it is best to focus on the essential underpinnings even of more advanced game theory: the three theoretical building blocks represented by the rules of the game, the expected utility model, and the concept of equilibrium.

This paper proposes a decomposition of the cost of risk (as measured by a risk premium) across intervals/quantiles of the payoff distribution. The analysis is based on general smooth risk preferences. While this includes the expected utility model as a special case, the investigation is done under a broad class of non-expected utility models. We decompose the risk premium into additive components across quantiles. Defining downside risk as the risk associated with a lower quantile, this provides a basis to evaluate the cost of exposure to downside risk. We derive a local measure of the cost of risk associated with each quantile. It establishes linkages between the cost of risk, risk preferences and the distribution of risky prospects across quantiles (as measured by quantile variance and skewness). The analysis gives new and useful information on how risk aversion, exposure to downside risk and departures from the expected utility model interact as they affect the risk premium.

One area of research concerning the demand for insurance focuses on determining the optimal form for the indemnification function. This literature shows that from the viewpoint of the insured the deductible form for the indemnification function is optimal under quite general circumstances both in the expected utility model and in nonexpected utility models that preserve second degree stochastic dominance.1 This result depends on the way in which deductible insurance is priced. How other forms of insurance are priced obviously matters if the deductible form is to be optimal. It is also the case that the pricing of deductible policies relative to one another plays a role in demonstrating the optimality of the deductible form.

In this chapter a new model of choice under risk within the framework of RDU theory is offered. The suggested model contains the expected utility model as a special case. Compared to anticipated utility theory it is more general in one respect and more restrictive in another. It is more general since it allows the probability distortion to depend on the prizes available. But it restricts on the other hand these distortions to be homogeneous in the probabilities.

This paper proposes a different point of view about the interpretation of EU models. It looks at the role of OR and AND aggregations in their construction and gives a fairly general setting for the study of these connectives. Finally, it examines which conditions may justify the traditional EU models and which novelties may derive from their weakening.

Preferences of Patients With HR+ and HER2− Breast Cancer Regarding Hormonal and Targeted Therapies in the First Line of Their Metastatic Stage: A Discrete Choice Experiment

An expected utility model is formulated and axiomatized which allows the decision maker to specify his actions in the form of partial rather than complete contingency plans, and to simultaneously choose goals and actions in end-mean pairs. Both utility and probability are conditioned on selected goals and actions, and both are defined over the same set of events. For adaptive sequential decision problems this symmetrical treatment of utility and probability permits the expected utility criterion function to be directly updated in each decision period via transitional utility assessments in a manner analogous to Bayes' rule for updating probability distributions.