Abstract
This paper introduces a definition of stochastic superiority. One random variable is stochastically superior to another whenever it stochastically dominates the other after the risk in each random variable has been optimally reduced. Stochastic superiority is implied by stochastic dominance, but the reverse is not true. Stochastic superiority allows more pairs of random alternatives to be ranked, and efficient sets to be smaller. A very strong sufficient condition for stochastic superiority is demonstrated to also be necessary when preferences are risk averse. This condition provides a relatively easy way to conduct stochastic superiority tests. As an alternative to “almost stochastic dominance,” stochastic superiority also provides a natural solution to the “left tail problem” that arises often when comparing random alternatives.
Similar content being viewed by others
Notes
Whereas the primary and secondary decisions are simultaneously made in real-world situations, the primary decision is typically made in isolation in a laboratory environment. Therefore, it is not surprising that the risk attitudes elicited based on choices made in the lab consistently fail to predict choices in the field (e.g., Charness et al. 2020).
Since \(\stackrel{{\sim }}{\text{x}}\) has a positive probability mass to the left of the entire distribution of \(\stackrel{{\sim }}{\text{y}}\), \(\stackrel{{\sim }}{\text{x}}\) does not stochastically dominate \(\stackrel{{\sim }}{\text{y}}\) in any degree. Of course, since \(\mathrm{E}(\stackrel{{\sim }}{\text{x}})\) > E(\(\stackrel{{\sim }}{\text{y}}\)), \(\stackrel{{\sim }}{\text{y}}\) does not stochastically dominate \(\stackrel{{\sim }}{\text{x}}\) in any degree either.
For a survey of this literature, see Barbera et al. (2004).
Levy (2006), Chapter 4.
In Levy and Kroll’s analysis, the sets {\(\stackrel{{\sim }}{\text{x}}\)i} and {\(\stackrel{{\sim }}{\text{y}}\)j} are the returns from portfolios that are composed of a risky asset \(\stackrel{{\sim }}{\text{x}}\) or \(\stackrel{{\sim }}{\text{y}}\), and a riskless asset. In this regard, their decision model is very similar to that presented here.
Sandmo (1971) uses these linear transformations in his analysis of the firm under price uncertainty. If α > 1, the transformation would not be risk reducing; if α < 0, the risk-reducing transformation would go too far to preserve any risk structure of the original random prospect.
Fishburn (1974) exploits this same thing when demonstrating convex stochastic dominance theorems.
Note that with the additional requirement that the minimum value of \({\stackrel{{\sim }}{\text{L}}}_{\text{i}}\) (i = 1 or 2) with a positive probability (density) be zero, the decomposition of \(\stackrel{{\sim }}{\text{x}}\) = wi − \({\stackrel{{\sim }}{\text{L}}}_{1}\) and \(\stackrel{{\sim }}{\text{y}}\) = w2 − \({\stackrel{{\sim }}{\text{L}}}_{2}\) must each be unique.
Levy (2006), Chapter 13, 331–332.
\(\stackrel{{\sim }}{\text{x}}\) does not stochastically dominate \(\stackrel{{\sim }}{\text{y}}\) in any degree because \(\stackrel{{\sim }}{\text{x}}\) has a positive probability mass to the left of the entire distribution of \(\stackrel{{\sim }}{\text{y}}\). \(\stackrel{{\sim }}{\text{y}}\) does not stochastically dominate \(\stackrel{{\sim }}{\text{x}}\) in any degree because \(\mathrm{E}(\stackrel{{\sim }}{\text{x}})\) > E( \(\stackrel{{\sim }}{\text{y}}\)).
This means that an α-portion of the loss is retained by the individual and the remaining (1 - α)-portion of the loss is transferred to the insurer.
References
Barbera, S., Bossert, W., & Pattanaik, P. K., et al. (2004). Ranking sets of objects. In S. Barbera (Ed.), Handbook of utility theory. Springer Science+Business Media, Inc.
Bayrak, O. K., & Hey, J. D. (2020). Decisions under risk: Dispersion and skewness. Journal of Risk and Uncertainty, 61(1), 1–24.
Chang, J.-R., Liu, W.-H., & Hung, M.-W. (2019). Revisiting generalized almost stochastic dominance. Annals of Operations Research, 281, 175–192.
Charness, G., Garcia, T., Offerman, T., & Villeval, M. C. (2020). Do measures of risk attitude in the laboratory predict behavior under risk in and outside of the laboratory? Journal of Risk and Uncertainty, 60(2), 99–123.
Denuit, M., Dhaene, J., Goovaerts, M., & Kaas, R. (2005). Actuarial theory for dependent risks: Measures, orders and models. Wiley.
Denuit, M., Huang, R. J., & Tzeng, L. Y. (2014). Bivariate almost stochastic dominance. Economic Theory, 57, 377–405.
Eeckhoudt, L., & Gollier, C. (2005). The impact of prudence on optimal prevention. Economic Theory, 26, 989–994.
Ehrlich, I., & Becker, G. (1972). Market insurance and self-insurance. Journal of Political Economy, 80, 623–648.
Fishburn, P. (1974). Convex stochastic dominance with continuous distribution functions. Journal of Economic Theory, 7, 143–158.
Guo, X., Zhu, X., Wong, W.-K., & Zhu, L. (2013). A note on almost stochastic dominance. Economics Letters, 121, 252–256.
Hadar, J., & Russell, W. (1969). Rules for ordering uncertain prospects. American Economic Review, 59, 25–34.
Hanoch, G., & Levy, H. (1969). The efficiency analysis of choices involving risk. Review of Economic Studies, 36, 335–346.
Leshno, M., & Levy, H. (2002). Preferred by “all” and preferred by “most” decision makers: Almost stochastic dominance. Management Science, 48, 1074–1085.
Levy, H. (2006). Stochastic dominance: Investment decision making under uncertainty (2nd ed.). Springer Science+Business Media Inc.
Levy, H., & Kroll, Y. (1976). Stochastic dominance with riskless assets. Journal of Financial and Quantitative Analysis, 11, 743–777.
Levy, H., & Kroll, Y. (1978). Ordering uncertain options with borrowing and lending. Journal of Finance, 33, 553–574.
Liu, L., & Meyer, J. (2021). Almost stochastic dominance: Magnitude constraints on risk aversion. Working paper, Private Enterprise Research Center, Texas A&M University. Retrieved from https://perc.tamu.edu/publication?id=248
Luo, C., & Tan, C. H. (2020). Almost stochastic dominance for most risk-averse decision makers. Decision Analysis, 17, 169–184.
Meyer, J. (1977a). Second degree stochastic dominance with respect to a function. International Economic Review, 18, 477–487.
Meyer, J. (1977b). Choice among distributions. Journal of Economic Theory, 14, 326–336.
Meyer, J. (1987). Two-moments decision models and expected utility maximization. American Economic Review, 77, 421–430.
Neilson, W., & Stowe, J. (2002). A further examination of cumulative prospect theory parameterizations. Journal of Risk and Uncertainty, 24(1), 31–46.
Quiggin, J. (2014). Non-expected utility models under objective uncertainty. In M. Machina & W. K. Viscusi (Eds.), Handbook of the economics of risk and uncertainty. North Holland Press.
Rothschild, M., & Stiglitz, J. (1970). Increasing risk I: A definition. Journal of Economic Theory, 2, 225–243.
Sandmo, A. (1971). On the theory of the competitive firm under price uncertainty. American Economic Review, 61, 65–73.
Shaked, M., & Shanthikumar, J. G. (2007). Stochastic orders. Springer.
Tsetlin, I., Winkler, R. L., Huang, R. J., & Tzeng, L. Y. (2015). Generalized almost stochastic dominance. Operations Research, 63, 363–377.
Tversky, A., & Kahneman, D. (1992). Advances in prospect theory: Cumulative representation of uncertainty. Journal of Risk and Uncertainty, 5(4), 297–323.
Tzeng, L. Y., Huang, R. J., & Shih, P.-T. (2013). Revisiting almost second-degree stochastic dominance. Management Science, 59, 1250–1254.
Viscusi, W. K. (1989). Prospective reference theory: Toward an explanation of the paradoxes. Journal of Risk and Uncertainty, 2(3), 235–263.
Wong, W.-K., & Ma, C. (2008). Preferences over location-scale family. Economic Theory, 37, 119–146.
Acknowledgements
We would like to thank an anonymous referee, Yoichiro Fujii, Michael Hoy, Rachel Huang, Lu Li, Henri Louberge, Alex Muermann, Larry Tzeng, Kit Pong “Keith” Wong, Lin Zhao, and participants of the 2019 EGRIE seminar for helpful comments and suggestions.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
1.1 Proof of Theorem 1
Assume \(\alpha \stackrel{{\sim }}{\text{x}}+\left(1-\alpha \right)\rho\) stochastically dominates \(\tilde{y}\) for all u(x) in U. Then from Lemma 1, for any λ in [0, 1] and \(\gamma = \left( {1 - \lambda } \right)\rho ,\,\lambda \left[ {\alpha \tilde{x} + \left( {1 - \alpha } \right)\rho } \right] + \left( {1 - \lambda } \right)\rho = \lambda \alpha \tilde{x} + \left( {1 - \lambda \alpha } \right)\rho\) stochastically dominates \(\lambda \tilde{y} + \left( {1 - \lambda } \right)\rho\) for all u(x) in U. Put differently, for every element in the set \(\left\{ {\lambda \tilde{y} + \left( {1 - \lambda } \right)\rho } \right\}\), there is an element in the set \(\left\{ {\lambda \alpha \tilde{x} + \left( {1 - \lambda \alpha } \right)\rho } \right\}\) that dominates the former for all u(x) in U. This is of course sufficient for the optimal choice from the set \(\left\{ {\alpha \tilde{x} + \left( {1 - \alpha } \right)\rho } \right\}\) to be preferred or indifferent to the optimal choice from the set \(\left\{ {\beta \tilde{y} + \left( {1 - \beta } \right)\rho } \right\}\) for every utility function in the set U. Therefore, \(\stackrel{{\sim }}{\text{x}}\) is stochastically superior to \(\tilde{y}\) for all u(x) in U for the given ρ. Q.E.D.
1.2 Proof of Theorem 2
Let \({\text{U}}\left( \alpha \right) = {\text{Eu}}\left( {\alpha \tilde{x} + \left( {1 - \alpha } \right)\rho } \right)\) for α in [0, 1] and a given ρ. What must be shown is that if for any u(x) in U there exists an α such that \({\text{U}}\left( \alpha \right) \ge {\text{Eu}}\left( {\tilde{y}} \right)\), then there exists an α0 in [0, 1] such that \({\text{U}}\left( {\alpha_{0} } \right) \ge {\text{Eu}}\left( {\tilde{y}} \right)\) for all u(x) in U. Define Au to be the set of all α in [0, 1] such that \({\text{U}}\left( \alpha \right) \ge {\text{Eu}}\left( {\tilde{y}} \right)\). For concave utility functions, Au is a closed line segment in [0, 1]. Let A* denote the intersection of these line segments. If A* is not empty, then any α0 in A* is such that \({\text{U}}\left( {\alpha_{0} } \right) \ge {\text{Eu}}\left( {\tilde{y}} \right)\) for all u(x) in U and we are done.
Assume instead that A* is empty. We must show that this is not possible when \(\stackrel{{\sim }}{\text{x}}\) is stochastically superior to \(\tilde{y}\). Let αu denote the maximum value for α such that \({\text{U}}\left( \alpha \right) \ge {\text{Eu}}\left( {\tilde{y}} \right)\) for utility function u(x). That is, αu is the right end point of the line segment Au. Denote this set of real numbers, one for each utility function, as \(\left\{ {\alpha_{{\text{u}}} } \right\}\). Let αL denote the infimum or greatest lower bound of this set \(\left\{ {\alpha_{{\text{u}}} } \right\}\). αL exists because \(\left\{ {\alpha_{{\text{u}}} } \right\}\) is a bounded set of real numbers. Similarly, let \(\alpha_{{\text{u}}}^{ * }\) denote the minimum value for α such that \({\text{U}}\left( \alpha \right) \ge {\text{Eu}}\left( {\tilde{y}} \right)\). \(\alpha_{{\text{u}}}^{ * }\) is the left end point of the line segment which is Au. Denote this set of real numbers \(\left\{ {\alpha_{{\text{u}}}^{ * } } \right\}\). Let αR denote the supremum or least upper bound of this set \(\left\{ {\alpha_{{\text{u}}}^{ * } } \right\}\).
If αL > αR, then for all α in [αR, αL], \({\text{U}}\left( \alpha \right) \ge {\text{Eu}}\left( {\tilde{y}} \right)\) for all u(x) in U and hence A* is not empty. When αL = αR, and αL is in \(\left\{ {\alpha_{{\text{u}}} } \right\}\) and αR is in \(\left\{ {\alpha_{{\text{u}}}^{ * } } \right\}\), the same thing is true; that is \({\text{U}}\left( {\alpha_{L} } \right) \ge {\text{Eu}}\left( {\tilde{y}} \right)\) for all u(x) in U. Thus, the hypothesis A* empty rules this out.
If αL < αR, for any α0 in [αL, αR], there exists a utility function u1(x) such that \({\text{U}}_{1} \left( {\alpha_{0} } \right) < {\text{Eu}}\left( {\tilde{y}} \right)\) and the α* which maximizes U1(α) occurs at a point α* < α0 (because αL < α0). Similarly, there also exists a utility function u2(x) such that \({\text{U}}_{2} \left( {\alpha_{0} } \right) < {\text{Eu}}\left( {\tilde{y}} \right)\) and the α* which maximizes U2(α) occurs at a point α* > α0 (because α0 < αR). This same thing is true when αL = αR, and αL is not in \(\left\{ {\alpha_{{\text{u}}} } \right\}\) and αR is not in \(\left\{ {\alpha_{{\text{u}}}^{ * } } \right\}\). Either of these two cases allows construction of a utility function of the form \({\text{u}}_{\lambda } \left( x \right) = \lambda {\text{u}}_{1} \left( x \right) + \left( {1 - \lambda } \right){\text{u}}_{2} \left( x \right)\) which satisfies \({\text{U}}_{\lambda } \left( {\alpha_{0} } \right) < {\text{Eu}}\left( {\tilde{y}} \right)\) and the maximum α* for this utility function occurs at α0. Hence stochastic superiority is violated so this case is also not possible.
The final two cases to discuss are when αL = αR, and αL is in \(\left\{ {\alpha_{{\text{u}}} } \right\}\) and αR is not in \(\left\{ {\alpha_{{\text{u}}}^{ * } } \right\}\), or vice versa. The discussion is for the first of these cases and the second is resolved in a similar fashion. When αL = αR, and αL is in \(\left\{ {\alpha_{{\text{u}}} } \right\}\) and αR is not in \(\left\{ {\alpha_{{\text{u}}}^{ * } } \right\}\), there exists a utility function u1(x) such that \({\text{U}}_{1} \left( {\alpha_{{\text{L}}} } \right) = {\text{Eu}}\left( {\tilde{y}} \right)\) and the α* which maximizes U1(α) occurs at a point α* < αL. There also exists a utility function u2(x) such that U2(αL) < Eu(\(\stackrel{{\sim }}{\text{y}}\)) and the α* which maximizes U2(α) occurs at a point α* > αL. This allows construction of a utility function of the form \({\text{u}}_{\lambda } \left( x \right) = \lambda {\text{u}}_{1} \left( x \right) + \left( {1 - \lambda } \right){\text{u}}_{2} \left( x \right)\) which satisfies \({\text{U}}_{\lambda } \left( {\alpha_{{\text{L}}} } \right) < {\text{Eu}}\left( {\tilde{y}} \right)\) and the maximum α* for this utility function occurs at αL. Thus, this case is not possible either and it can be concluded that A* is not empty when stochastic superiority of \(\stackrel{{\sim }}{\text{x}}\) over \(\tilde{y}\) holds. Q.E.D.
1.3 Proof of Theorem 3
Suppose that \(\tilde{x}\) is stochastically superior to \(\tilde{y}\) on some convex set of concave utility functions U for a given ρ. Then, from Theorem 2, there exists an α0 in [0, 1] such that \(\alpha_{0} \tilde{x} + \left( {1 - \alpha_{0} } \right)\rho\) stochastically dominates \(\tilde{y}\) on U. Since for \(\rho^{\prime} \ge \rho ,\,\alpha_{0} \tilde{x} + \left( {1 - \alpha_{0} } \right)\rho^{\prime}\) stochastically dominates \(\alpha_{0} \tilde{x} + \left( {1 - \alpha_{0} } \right)\rho\), by transitivity it also stochastically dominates \(\tilde{y}\) on U. Therefore, according to Theorem 1, \(\stackrel{{\sim }}{\text{x}}\) is stochastically superior to \(\tilde{y}\) on U for ρ′. Q.E.D.
1.4 Proof of Theorem 4
Sufficiency is obvious. To show necessity, suppose that \(\left(\stackrel{{\sim }}{\text{x}}, {\rho }_{x}\right)\) is stochastically superior to \(\left(\stackrel{{\sim }}{\text{y}}, {\rho }_{y}\right)\) on U for some ρy such that ρx ≥ ρy. Then for every u(x) in U there exists an α ∈ [0, 1] such that \(\alpha \tilde{x} + \left( {1 - \alpha } \right)\rho_{x}\) is preferred or indifferent to \(\tilde{y}\). According to Theorem 2, there exists an α0 ∈ [0, 1] such that \(\alpha_{0} \tilde{x} + \left( {1 - \alpha_{0} } \right)\rho_{x}\) stochastically dominates \(\tilde{y}\) on U. From Theorem 1, this implies that \(\left(\stackrel{{\sim }}{\text{x}}, {\rho }_{x}\right)\) is stochastically superior to \(\left(\stackrel{{\sim }}{\text{y}}, {\rho }_{x}\right)\) on U. Q.E.D.
1.5 Proof of Theorem 5
Equation 1 implies that ρx > ρy, and therefore according to Theorems 1, 2 and 4, \(\left(\stackrel{{\sim }}{\text{x}}, {\rho }_{x}\right)\) is stochastically superior to \(\left(\stackrel{{\sim }}{\text{y}}, {\rho }_{y}\right)\) in the second degree if and only if there is an α in [0, 1] such that \(\alpha \tilde{x} + \left( {1 - \alpha } \right)\rho_{x}\) stochastically dominates \(\tilde{y}\) in the second degree, where \(\rho_{x} = {\text{w}} - {\text{e}}_{2} - \left( {1 + \pi } \right){\text{p}}_{2} {\text{L}}\).
To establish the condition under which there exists an α in [0, 1] such that \(\alpha \tilde{x} + \left( {1 - \alpha } \right)\rho_{x}\) stochastically dominates \(\tilde{y}\) in the second degree, note that \(\alpha \tilde{x} + \left( {1 - \alpha } \right)\rho_{x}\) is a binary distribution yielding a low outcome \(\alpha \left( {{\text{w}} - {\text{e}}_{2} - {\text{L}}} \right) + \left( {1 - \alpha } \right)\left( {{\text{w}} - {\text{e}}_{2} - \left( {1 + \pi } \right){\text{p}}_{2} {\text{L}}} \right)\) and a high outcome \(\alpha \left( {{\text{w}} - {\text{e}}_{2} } \right) + \left( {1 - \alpha } \right)\left( {{\text{w}} - {\text{e}}_{2} - \left( {1 + \pi } \right){\text{p}}_{2} {\text{L}}} \right)\) with probabilities p2 and (1 − p2) respectively. Similarly, \(\tilde{y}\) is a binary distribution yielding a low outcome \({\text{w}} - {\text{e}}_{1} - {\text{L}}\) and a high outcome w − e1 with probabilities p1 and (1 − p1) respectively. Because p2 < p1, \(\alpha \tilde{x} + \left( {1 - \alpha } \right)\rho_{x}\) stochastically dominates \(\tilde{y}\) in the second degree if and only if the low outcome of the former is no less than the low outcome of the latter, and the mean of the former is no less than the mean of the latter. These conditions are expressed as eqs. (3) and (4) given below:
and
Now Eqs. (3) and (4) can be equivalently expressed as:
and
respectively.
There exists such an α that satisfies both Eqs. (5) and (6) if and only if
or equivalently
Rights and permissions
About this article
Cite this article
Liu, L., Meyer, J. Stochastic superiority. J Risk Uncertain 62, 225–246 (2021). https://doi.org/10.1007/s11166-021-09362-9
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11166-021-09362-9