Ilia Tsetlin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Ilia Tsetlin has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Robert L. Winkler
Fuqua School of Business, Duke University, Durham, North Carolina 27708
8 shared publications - Michel Regenwetter
Department of Psychology, University of Illinois at Urbana-Champaign
3 shared publications - Saša Pekeč
Fuqua School of Business, Duke University, Durham, North Carolina 27708
3 shared publications - Alfred Müller
Department Mathematik, Universität Siegen, 57072 Siegen, Germany
2 shared publications - Anil Gaba
Department of Decision Sciences. INSEAD. Singapore
2 shared publications - Marco Scarsini
Dipartimento di Economia e Finanza, Luiss University, 00197 Roma, Italy
2 shared publications
Latest Publications
- 15 May 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
Classical Lottery in Action: Quantifying Risk and Evaluating Uncertainty
The rarity of objectively known probabilities undermines the risk-ambiguity dichotomy, challenging the practical relevance of related theories. We return to classical lotteries—coins, dice, and similar devices—which inspired early probability theories through the idea of equiprobable outcomes and are widely considered strong candidates for objective probability. We adapt axioms of expected utility for risk to advocate average utility for classical lotteries, highlighting their conceptual affinity. Any general unknown event is conceived as a collection of possible classical-lottery frequencies, consistent with the ambiguity literature, and we suggest normative principles for their aggregation into an indifferent matching frequency. These principles identify a model in the spirit of the smooth model of ambiguity; the agent assigns subjective probabilities over the frequencies and aggregates them via a nonlinear mixture. The nonlinearity reflects the agent’s distinct attitudes toward classical-lottery frequencies versus subjective probabilities; a concave mixture, for example, captures ambiguity aversion. Our approach provides a concrete justification for the distinct attitudes and presents several advantages for model elicitation. We illustrate the theory’s applicability through examples and, especially, social policy evaluation where the veil of ignorance can be seen as a classical lottery. This paper was accepted by Manel Baucells, behavioral economics and decision analysis. Funding: J. Li was supported by the General Research Fund of the Hong Kong Research [Grant Council under research project LU13500322]. I. Tsetlin and F. Wang do not have fundings that need to be disclosed.
- 15 May 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
Classical Lottery in Action: Quantifying Risk and Evaluating Uncertainty
The rarity of objectively known probabilities undermines the risk-ambiguity dichotomy, challenging the practical relevance of related theories. We return to classical lotteries—coins, dice, and similar devices—which inspired early probability theories through the idea of equiprobable outcomes and are widely considered strong candidates for objective probability. We adapt axioms of expected utility for risk to advocate average utility for classical lotteries, highlighting their conceptual affinity. Any general unknown event is conceived as a collection of possible classical-lottery frequencies, consistent with the ambiguity literature, and we suggest normative principles for their aggregation into an indifferent matching frequency. These principles identify a model in the spirit of the smooth model of ambiguity; the agent assigns subjective probabilities over the frequencies and aggregates them via a nonlinear mixture. The nonlinearity reflects the agent’s distinct attitudes toward classical-lottery frequencies versus subjective probabilities; a concave mixture, for example, captures ambiguity aversion. Our approach provides a concrete justification for the distinct attitudes and presents several advantages for model elicitation. We illustrate the theory’s applicability through examples and, especially, social policy evaluation where the veil of ignorance can be seen as a classical lottery. This paper was accepted by Manel Baucells, behavioral economics and decision analysis. Funding: J. Li was supported by the General Research Fund of the Hong Kong Research [Grant Council under research project LU13500322]. I. Tsetlin and F. Wang do not have fundings that need to be disclosed.
- 12 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
Frontiers in Operations: You, Me, or We? Coproductive Principal-Agent Dynamics
Problem definition: Projects are often initiated by a single person—a principal—who then decides whether to form a team by sharing project value with an agent. We ask the following: When does a principal form a team, and which operating mode emerges in equilibrium—single execution by the principal, delegated execution to the agent, or joint execution? Methodology/results: We consider a coproductive principal-agent model with endogenous team formation. In the second best, joint execution is less frequent than optimal, whereas both single and delegated execution are chosen too much. With a linear contract, common with nonfinancial output (credits, coauthorship) or in entrepreneurial ventures, there is too much single execution; that is, the principal does not form a team often enough, and there is either too little joint execution when one of the workers has low productivity or too little delegation otherwise. Overall, the inefficiency in project execution (due to moral hazard) appears less severe than the inefficiency in the principal’s team formation decision (due to project hoarding). Managerial implications: Managerial under-delegation fundamentally stems from project hoarding; that is, principals do not partner enough. When principals choose to form a team, they then might delegate too much; paradoxically, this happens when agents have low productivity. Although the existing literature focuses on eliciting agent effort for a given operating mode, our analysis suggests the decision to form a team (or not) is a more critical issue. History: This paper has been accepted in the Manufacturing & Service Operations Management Frontiers in Operations Initiative. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2025.0016 .
- 12 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
Frontiers in Operations: You, Me, or We? Coproductive Principal-Agent Dynamics
Problem definition: Projects are often initiated by a single person—a principal—who then decides whether to form a team by sharing project value with an agent. We ask the following: When does a principal form a team, and which operating mode emerges in equilibrium—single execution by the principal, delegated execution to the agent, or joint execution? Methodology/results: We consider a coproductive principal-agent model with endogenous team formation. In the second best, joint execution is less frequent than optimal, whereas both single and delegated execution are chosen too much. With a linear contract, common with nonfinancial output (credits, coauthorship) or in entrepreneurial ventures, there is too much single execution; that is, the principal does not form a team often enough, and there is either too little joint execution when one of the workers has low productivity or too little delegation otherwise. Overall, the inefficiency in project execution (due to moral hazard) appears less severe than the inefficiency in the principal’s team formation decision (due to project hoarding). Managerial implications: Managerial under-delegation fundamentally stems from project hoarding; that is, principals do not partner enough. When principals choose to form a team, they then might delegate too much; paradoxically, this happens when agents have low productivity. Although the existing literature focuses on eliciting agent effort for a given operating mode, our analysis suggests the decision to form a team (or not) is a more critical issue. History: This paper has been accepted in the Manufacturing & Service Operations Management Frontiers in Operations Initiative. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2025.0016 .
- Open Access
- 14 August 2025
- Published by University of Michigan Library in Philosophers' Imprint
- Vol. 25
Resolute and Correlated Bayesians
This paper suggests a new normative approach for combining beliefs. We call it the evidence-first method. Instead of aggregating credences alone, as the prevailing approaches, we focus instead on eliciting a group’s full probability distribution on the basis of the evidence available to its members. This is an altogether different way of combining beliefs. The method has four main benefits: (1) it captures the weight, or resilience, of a group’s belief; (2) it is sensitive to correlation among its individuals; (3) it is commutative under updating; and (4) it can be seen as a generalization of weighted averaging and likelihood ratio approaches. More broadly, it encourages an overall rethinking of the belief combination problem away from aggregating bare credences and toward appropriately combining evidence.
- 14 August 2025
- Published by University of Michigan Library in Philosophers' Imprint
- Vol. 25
Open AccessResolute and Correlated Bayesians
This paper suggests a new normative approach for combining beliefs. We call it the evidence-first method. Instead of aggregating credences alone, as the prevailing approaches, we focus instead on eliciting a group’s full probability distribution on the basis of the evidence available to its members. This is an altogether different way of combining beliefs. The method has four main benefits: (1) it captures the weight, or resilience, of a group’s belief; (2) it is sensitive to correlation among its individuals; (3) it is commutative under updating; and (4) it can be seen as a generalization of weighted averaging and likelihood ratio approaches. More broadly, it encourages an overall rethinking of the belief combination problem away from aggregating bare credences and toward appropriately combining evidence.
- 1 March 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (2) , 879-893
Multivariate Almost Stochastic Dominance: Transfer Characterizations and Sufficient Conditions Under Dependence Uncertainty
Decision with Several Objectives Under Uncertainty Important decisions typically involve multiple objectives and uncertainty. Assessing both multivariate utility and multivariate distributions for the attributes can be challenging. Moreover, big decisions are usually made by boards or committees with members holding divergent views and preferences and facing pressures from different stakeholders. Thus, a full-blown traditional decision analysis that leads to the computation of expected utility is very difficult at best and often not possible. In “Multivariate Almost Stochastic Dominance: Transfer Characterizations and Sufficient Conditions Under Dependence Uncertainty” Müller, Scarsini, Tsetlin, and Winkler develop sufficient conditions for multivariate almost stochastic dominance based on marginal distributions of the attributes or just on their means and variances. Such tools, consistent with normative decision analysis, are useful when making important decisions in today’s fast-moving and often complex world.
- 1 March 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (2) , 879-893
Multivariate Almost Stochastic Dominance: Transfer Characterizations and Sufficient Conditions Under Dependence Uncertainty
Decision with Several Objectives Under Uncertainty Important decisions typically involve multiple objectives and uncertainty. Assessing both multivariate utility and multivariate distributions for the attributes can be challenging. Moreover, big decisions are usually made by boards or committees with members holding divergent views and preferences and facing pressures from different stakeholders. Thus, a full-blown traditional decision analysis that leads to the computation of expected utility is very difficult at best and often not possible. In “Multivariate Almost Stochastic Dominance: Transfer Characterizations and Sufficient Conditions Under Dependence Uncertainty” Müller, Scarsini, Tsetlin, and Winkler develop sufficient conditions for multivariate almost stochastic dominance based on marginal distributions of the attributes or just on their means and variances. Such tools, consistent with normative decision analysis, are useful when making important decisions in today’s fast-moving and often complex world.
- 1 January 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (1) , 461-482
The When and How of Delegated Search
Contract Design for Outsourcing Search Firms commonly outsource search for new employees, real estate, or technology to external agents. What should the contracts with these agents look like, and under which conditions should companies even hire agents (as opposed to doing the search in house)? These are the questions studied in “The when and how of delegated search” by Zorc et al. The authors find that the optimal contracts pay the agent a per-time fee as well as a bonus for finding an acceptable alternative. The size of this bonus is defined on signing of the contract and decreases over time. The decision of whether to outsource at all hinges on the firm’s trade-off between speed and quality; in-house search becomes optimal for a firm that prioritizes quality, but outsourcing offers better speed.
- 1 January 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (1) , 461-482
The When and How of Delegated Search
Contract Design for Outsourcing Search Firms commonly outsource search for new employees, real estate, or technology to external agents. What should the contracts with these agents look like, and under which conditions should companies even hire agents (as opposed to doing the search in house)? These are the questions studied in “The when and how of delegated search” by Zorc et al. The authors find that the optimal contracts pay the agent a per-time fee as well as a bonus for finding an acceptable alternative. The size of this bonus is defined on signing of the contract and decreases over time. The decision of whether to outsource at all hinges on the firm’s trade-off between speed and quality; in-house search becomes optimal for a firm that prioritizes quality, but outsourcing offers better speed.
- 13 March 2024
- Published by University of Chicago Press in The British Journal for the Philosophy of Science
- Vol. 75 (1) , 000
Normativity, Epistemic Rationality, and Noisy Statistical Evidence
- 13 March 2024
- Published by University of Chicago Press in The British Journal for the Philosophy of Science
- Vol. 75 (1) , 000
Normativity, Epistemic Rationality, and Noisy Statistical Evidence
- 1 September 2022
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 70 (5) , 2851-2859
Technical Note—Ranking Distributions When Only Means and Variances Are Known
In “Technical Note—Ranking Distributions When Only Means and Variances Are Known,” Müller, Scarsini, Tsetlin, and Winkler address the question of ranking distributions when only the first two moments—that is, means and variances—are known. This is important in decision making under uncertainty, with potential applications in economics, finance, statistics, and other areas. Previous results require some assumptions about the shape of the distributions, while this paper’s approach is to impose bounds on how much marginal utility can change, thus constraining risk preferences. Such a ranking is consistent with almost stochastic dominance and provides a new connection between the Sharpe and Omega ratios from finance.
- 1 September 2022
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 70 (5) , 2851-2859
Technical Note—Ranking Distributions When Only Means and Variances Are Known
In “Technical Note—Ranking Distributions When Only Means and Variances Are Known,” Müller, Scarsini, Tsetlin, and Winkler address the question of ranking distributions when only the first two moments—that is, means and variances—are known. This is important in decision making under uncertainty, with potential applications in economics, finance, statistics, and other areas. Previous results require some assumptions about the shape of the distributions, while this paper’s approach is to impose bounds on how much marginal utility can change, thus constraining risk preferences. Such a ranking is consistent with almost stochastic dominance and provides a new connection between the Sharpe and Omega ratios from finance.
- 1 May 2020
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 68 (3) , 927-948
Deadlines, Offer Timing, and the Search for Alternatives
In “Deadlines, Timing, and the Search for Alternatives,” S. Zorc and I. Tsetlin consider decentralized matching markets like the job market for MBAs. They show that exploding offers (ones with a very short deadline) are often optimal. The benefit of those offers is to prevent the other party’s search for alternatives. However, the main argument in favor of longer deadlines is that people who already hold an offer become more selective when deciding whether to accept other alternatives (the acceptance deterrence effect). Thus, the goal of curtailing search can also be accomplished through increasing the other party’s selectivity instead of preventing them from searching altogether. A practical prescription for offer makers is to use exploding offers when acceptance deterrence is weak (e.g., when the other party’s alternatives will not change) and longer deadlines when it is strong (e.g., when you are able to react to the other party’s other offers).
- 1 May 2020
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 68 (3) , 927-948
Deadlines, Offer Timing, and the Search for Alternatives
In “Deadlines, Timing, and the Search for Alternatives,” S. Zorc and I. Tsetlin consider decentralized matching markets like the job market for MBAs. They show that exploding offers (ones with a very short deadline) are often optimal. The benefit of those offers is to prevent the other party’s search for alternatives. However, the main argument in favor of longer deadlines is that people who already hold an offer become more selective when deciding whether to accept other alternatives (the acceptance deterrence effect). Thus, the goal of curtailing search can also be accomplished through increasing the other party’s selectivity instead of preventing them from searching altogether. A practical prescription for offer makers is to use exploding offers when acceptance deterrence is weak (e.g., when the other party’s alternatives will not change) and longer deadlines when it is strong (e.g., when you are able to react to the other party’s other offers).
- 1 June 2018
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 66 (3) , 799-813
When to Abandon a Research Project and Search for a New One
We investigate the cost of the opportunity delayed by working on one project with uncertain success rather than searching for a new project. We answer this question: How long should a firm work on a research project with uncertain success before abandoning it if the only alternative is to search for a new project to work on? Rather than treating the opportunity as an exogenous alternative, this approach endogenizes the opportunity value and the attendant cost of its delay. We consider cases with both a finite and an infinite number of potential projects. We derive the optimal stopping time and show that it increases with the discount rate, with the rate of arrival of project failure, and with the prior probability that the project can be successfully completed. The optimal stopping time decreases with the rates of new project arrival and project completion.
- 1 June 2018
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 66 (3) , 799-813
When to Abandon a Research Project and Search for a New One
We investigate the cost of the opportunity delayed by working on one project with uncertain success rather than searching for a new project. We answer this question: How long should a firm work on a research project with uncertain success before abandoning it if the only alternative is to search for a new project to work on? Rather than treating the opportunity as an exogenous alternative, this approach endogenizes the opportunity value and the attendant cost of its delay. We consider cases with both a finite and an infinite number of potential projects. We derive the optimal stopping time and show that it increases with the discount rate, with the rate of arrival of project failure, and with the prior probability that the project can be successfully completed. The optimal stopping time decreases with the rates of new project arrival and project completion.
- 23 August 2017
- Published by Wiley in Journal of Risk and Insurance
- Vol. 85 (2) , 431-445
Multivariate Almost Stochastic Dominance
Almost stochastic dominance allows small violations of stochastic dominance rules to avoid situations where most decision makers prefer one alternative to another but stochastic dominance cannot rank them. We present the concepts of multivariate almost stochastic dominance and multivariate almost nth‐degree risk and their connections with a preference for combining good with bad. Then, we show how a preference for combining good with bad can be applied to obtain various comparative statics results, and we extend our approach to risk‐prone (convex) stochastic dominance, which relates to the opposite preference, for combining good with good and bad with bad.
- 23 August 2017
- Published by Wiley in Journal of Risk and Insurance
- Vol. 85 (2) , 431-445
Multivariate Almost Stochastic Dominance
Almost stochastic dominance allows small violations of stochastic dominance rules to avoid situations where most decision makers prefer one alternative to another but stochastic dominance cannot rank them. We present the concepts of multivariate almost stochastic dominance and multivariate almost nth‐degree risk and their connections with a preference for combining good with bad. Then, we show how a preference for combining good with bad can be applied to obtain various comparative statics results, and we extend our approach to risk‐prone (convex) stochastic dominance, which relates to the opposite preference, for combining good with good and bad with bad.