Testing theories with learnable and predictive representations

Al-Najjar, Nabil I.; Sandroni, Alvaro; Smorodinsky, Rann; Weinstein, Jonathan

doi:10.1016/j.jet.2010.07.003

Cited by 20 publications

(32 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some recent work by Olszewski and Sandroni [18] and Al-Najjar et al [1] is the most related to the current paper, in the sense that they assume a structure on the set of possible distributions in forecasting settings. They follow two earlier papers ( [6,17]) that show a non-manipulable forecasting test.…”

Section: Related Workmentioning

confidence: 97%

“…Al-Najjar et al [1] were also concerned by the delayed decision of existing tests and thus they also restricted the class of possible probabilistic processes. They presented a convex set P * of theories referred to as learnable and predictable; they showed that when the possible theories are drawn from this set then experts can be tested in a timely manner.…”

Section: Related Workmentioning

confidence: 99%

“…1 Assume that if Bob is informed then he knows the probability p * exactly 2 , while if Our Approach In this paper we attempt to explain the difficulty in screening experts and to characterize the cases where one can motivate informed experts without rewarding uninformed experts. We consider a scenario where the probability p * belongs to a set of distributions P, where P is not necessarily the set of all distributions but rather possibly a subset of it.…”

Section: Introductionmentioning

confidence: 99%

“…For simplicity of the exposition in the introduction, unless explicitly stated otherwise, we assume that an informed expert is perfectly informed. 3 In the context of forecast testing, restrictions on the set of distributions were recently considered by Olszewski and Sandroni [18] and Al-Najjar et al [1], see Section 1.3 for discussion of related work.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Only valuable experts can be valued

Babaioff

Blumrosen

Lambert

et al. 2011

Proceedings of the 12th ACM Conference on Electronic Commerce

View full text Add to dashboard Cite

Suppose a principal Alice wishes to reduce her uncertainty regarding some future payoff. Consider a self-proclaimed expert Bob that may either be an informed expert knowing an exact (or approximate) distribution of a future random outcome that may affect Alice's utility, or an uninformed expert who knows nothing more than Alice does. Alice would like to hire Bob and solicit his signal. Her goal is to incentivize an informed expert to accept the contract and reveal his knowledge while deterring an uninformed expert from accepting the contract altogether. The starting point of this work is a powerful negative result (Olszewski and Sandroni, 2007), which tells us that in the general case for any contract which guarantees an informed expert some positive payoff an uninformed expert (with no extra knowledge) has a strategy which guarantees him a positive payoff as well.At the face of this negative result, we reexamine the notion of an expert and conclude that knowing some hidden variable (i.e., the description of the aforementioned distribution), does not make Bob an expert, or at least not a "valuable expert". The premise of our paper is that if Alice only tries to incentivize experts which are valuable to her decision making then she can indeed screen them from uninformed experts.On a more technical level, we consider the case where Bob's signal about the distribution of a future event cannot be an arbitrary distribution but rather comes from some subset P of all possible distributions. We give rather tight conditions on P (which relate to its convexity), under which screening is possible. We formalize our intuition that if these conditions are not met then an expert is not guaranteed to be valuable. We give natural and arguably useful scenarios where indeed such a restriction on the distribution arise.

show abstract

Section: Related Workmentioning

confidence: 97%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Only valuable experts can be valued

Babaioff

Blumrosen

Lambert

et al. 2011

Proceedings of the 12th ACM Conference on Electronic Commerce

View full text Add to dashboard Cite

show abstract

“…Al-Najjar, Sandroni, Smorodinsky, and Weinstein (2010) propose an alternative paradigm in which effective testing is possible. Fortnow and Vohra (2009) take a different approach, obtaining positive results by restricting the class of strategies available to the false expert.…”

Section: Introductionmentioning

confidence: 99%

Nonmanipulable Bayesian testing

Stewart

2011

Journal of Economic Theory

View full text Add to dashboard Cite

This paper considers the problem of testing an expert who makes probabilistic forecasts about the outcomes of a stochastic process. I show that, as long as uninformed experts do not learn the correct forecasts too quickly, a likelihood test can distinguish informed from uninformed experts with high prior probability. The test rejects informed experts on some data-generating processes; however, the set of such processes is topologically small. These results contrast sharply with many negative results in the literature.

show abstract

A Cardinal Comparison of Experts

Kavaler

Smorodinsky

2020

Web and Internet Economics

View full text Add to dashboard Cite

Testing theories with learnable and predictive representations

Cited by 20 publications

References 17 publications

Only valuable experts can be valued

Only valuable experts can be valued

Nonmanipulable Bayesian testing

A Cardinal Comparison of Experts

Contact Info

Product

Resources

About