The Troika paradox
Highlights
► We present a new paradox for decision making under risk. ► The new paradox challenges many popular non-expected utility models. ► Experimental evidence on the new paradox is presented.
Section snippets
Falsified theories
Consider a binary lottery which yields outcome with probability and outcome with probability 1-. Suppose that a decision maker finds outcome to be at least as good as outcome . Let and denote the utility of outcome and outcome respectively. Many theories of decision under risk postulate that the utility of lottery is given by formula (1).
A non-decreasing function satisfies , . This function differs across
Experiment
In the experiment, subjects faced three binary choice problems described in the introduction. The probability information was conveyed through a composition of red and black cards. Fig. 1 shows the first binary choice problem as it was displayed in the experiment (translated from German).
The experiment was conducted at the University of Innsbruck (Austria). Altogether, 35 undergraduate students took part in two experimental sessions, which were conducted on the same afternoon (October 14,
Conclusion
The paradox presented in this paper combines old empirical facts into a new puzzle. People simultaneously exhibit risk aversion when facing probable gains and risk seeking when facing small-probability gains. Friedman and Savage (1948) previously tried to explain this fact. It was also popularized by Tversky and Kahneman (1992) as an ingredient of the fourfold pattern of risk attitudes.
Equally well-known is another empirical fact. When facing lotteries of a similar expected value, people often
References (28)
Nontransitive measurable utility
Journal of Mathematical Psychology
(1982)Subjectively weighted utility: a descriptive extension of the expected utility model
Organizational Behavior and Human Performance
(1978)Regret in decision making under uncertainty
Operations Research
(1982)Reverse common ratio effect
Journal of Risk and Uncertainty
(2010)Modifying the mean-variance approach to avoid violations of stochastic dominance
Management Science
(2010)- et al.
Imprecision as an account of the preference reversal phenomenon
American Economic Review
(2007) A generalization of the quasilinear mean with applications to the measurement of income inequality and decision theory resolving the Allais paradox
Econometrica
(1983)- et al.
Mixture symmetry and quadratic utility
Econometrica
(1991) The prediction of decisions among bets
Journal of Experimental Psychology
(1955)Elements of Psychophysics
(1860)
The utility analysis of choices involving risk
Journal of Political Economy
A theory of disappointment aversion
Econometrica
Risk, probability, and a new theory of cardinal utility
Journal of Political Economy
The predictive utility of generalized expected utility theories
Econometrica
Cited by (8)
Probability weighting and L-moments
2016, European Journal of Operational ResearchAxiomatization of weighted (separable) utility
2014, Journal of Mathematical EconomicsCitation Excerpt :An improved understanding of the descriptive properties of weighted utility then calls for empirical tests of bi-separable transitivity, which are rather scarce in the literature (see, however Blavatskyy, 2012).
Compound invariance implies prospect theory for simple prospects
2013, Journal of Mathematical PsychologyCitation Excerpt :For example, Prelec’s (1998) characterization of compound invariance probability weighting under prospect theory becomes the same mathematical theorem as the classical characterizations of CARA (constant absolute risk averse, or linear–exponential) and CRRA (constant relative risk averse, or log-power) utility under expected utility. Other “identical” theorems concern conditionally invariant probability weighting (Prelec, 1998) and the constant relative decreasing impatient (renamed unit invariant in this paper) time discount functions of Bleichrodt, Rohde, and Wakker (2009) and Ebert and Prelec (2007). Thus we provide the simplest proofs of these results presently available in the literature, and we extend them to domains more general than considered before.
A synopsis of mean-multiplicative deviation and risk dispersion theory
2017, 67th Annual Conference and Expo of the Institute of Industrial Engineers 2017