Recent developments in modeling preferences: Uncertainty and ambiguity

Abstract

In subjective expected utility (SEU), the decision weights people attach to events are their beliefs about the likelihood of events. Much empirical evidence, inspired by Ellsberg (1961) and others, shows that people prefer to bet on events they know more about, even when their beliefs are held constant. (They are averse to ambiguity, or uncertainty about probability.) We review evidence, recent theoretical explanations, and applications of research on ambiguity and SEU.In the last 40 years the leading theories of choice in economics and psychology have been the expected utility (EU) theory of von Neumann and Morgenstern (1947) and the subjective expected utility (SEU) theory of Savage (1954). Empirical violations have led to reexaminations of both kinds of theory. In Weber and Camerer (1987), we reviewed the evidence, axioms, and application of alternatives to EU. Here we do the same for SEU.EU assumes that the probabilities of outcomes are known. If preferences follow a set of simple axioms, they can be represented by a real-valued utility function--preferred choices have higher utility numbers--and the utility of a choice is the expected utility of its possible outcomes, weighted by their probabilities.In SEU, probabilities are not necessarily objectively known, so SEU applies more widely than EU. (Indeed, it is hard to think of an important natural decision for which probabilities are objectively known.) In SEU, decision makers choose acts, which have consequences that depend on which of several uncertain "states" occurs. People are *Thanks to

From anomalies to forecasts: Toward a descriptive model of decisions under risk, under ambiguity, and from experience.

“…Problem 23 shows a strong tendency to avoid the ambiguous option when gaining in the non-ambiguous option is associated with high probability (ambiguity rate of 15%, SD = 0.30). Both rates are significantly different from 0.5, t(124) = 12.0 and −13.1, both p < .001, and are in line with previous findings of studies in decisions in uncertain settings without feedback (e.g., Camerer & Weber, 1992).…”

Risky business: the neuroeconomics of decision making under uncertainty

“…To account for these uncertainty-induced deviations from expected utility models, Tversky and Kahneman proposed prospect theory 7,8 , which posits separate functions for how people judge probabilities and how they convert objective value to subjective utility. Other theorists have developed models that account for the effects of ambiguity on choice, by treating ambiguity as a distribution of probabilities (and thus converting it to risk) 9 or by modeling the psychological biases that ambiguity induces (such as attention to extreme outcomes) 10,11 .…”

Why environmental and resource economists should care about non-expected utility models

“…To distinguish these, we use the term event risk to refer to the usual uncertainty about an event or outcome (e.g., the risk of getting cancer from a substance in drinking water) and the term ambiguity to refer to uncertainty about the risk itself (e.g., the uncertainty about what the actual probability of getting cancer is). In reviewing the literature on ambiguity, Camerer and Weber (1992) found no correlation between risk attitudes and ambiguity attitudes. They also found the existence of ''substantial premiums to avoid ambiguity -around 10-20% of expected value or expected probability'' (p. 340).…”

“…However, as Ellsberg (1961) showed experimentally, individuals do not follow this reductionist procedure; ambiguity does affect preferences. Camerer and Weber (1992) describe the nature of the ambiguity problem as follows: Suppose you flip a coin 1000 times and get 500 heads. Then take a second coin and flip it twice yielding one head and one tail.…”

“…The PWF for environmental risks is (to our knowledge) an unexplored empirical question. 9 Readers are referred to Kelsey and Quiggin (1992), Camerer and Weber (1992), Quiggin (1993) and Harless and Camerer (1994) for much broader reviews of the literature and perspectives on the relative utility of different models. 10 The RDEU model has, however, also been applied to conditions of ambiguity by Segal (1987), who uses the RDEU to apply to second-order probability distributions.…”