A Category Test that Cannot be Manipulated∗
نویسندگان
چکیده
In Dekel and Feinberg (2004) we suggested a test for discovering whether a potential expert is informed of the distribution of a stochastic process. This category test requires predicting a “small”– category I – set of outcomes. In this paper we show that under the continuum hypothesis there is a category test that cannot be manipulated, i.e. such that no matter how the potential expert randomizes his prediction, there will be realizations where he will fail to pass the test with probability 1. The set of realizations where he fails can be made large – a category II set. Moreover, this result holds for the finite approximations of the category test where the non-expert is failed in finite time and the expert is failed with small probability. JEL Classification: K9
منابع مشابه
No Manipulation Results for Non-Bayesian Tests
In Dekel and Feinberg (2004) we suggested a test for discovering whether a potential expert is informed of the distribution of a stochastic process. This category test requires predicting a “small”– category I – set of outcomes. In this paper we show that there is a randomized category test that cannot be manipulated, i.e. such that no matter how the potential expert randomizes his prediction, ...
متن کاملNo . 1957 Testing Multiple Forecasters
We consider a cross-calibration test of predictions by multiple potential experts in a stochastic environment. This test checks whether each expert is calibrated conditional on the predictions made by other experts. We show that this test is good in the sense that a true expert—one informed of the true distribution of the process—is guaranteed to pass the test no matter what the other potential...
متن کاملTask Influences on Category Learning
Two plausible influences on category learning are presentation format and the learner’s beliefs about future category uses. Standard experimental designs typically do not manipulate these two dimensions independently, and so their effects cannot be easily disentangled, particularly since both plausibly affect task difficulty. In the present paper, we independently manipulated the two influences...
متن کاملTesting Multiple Forecasters∗
We consider a cross-calibration test of predictions by multiple potential experts in a stochastic environment. This test checks whether each expert is calibrated conditional on the predictions made by other experts. We show that this test is good in the sense that a true expert—one informed of the true distribution of the process—is guaranteed to pass the test no matter what the other potential...
متن کاملShort vs. Long Category Perception Affected by Vowel Quality
The paper studies the impact of intrinsic vowel duration on category perception in boundary conditions by examining the short vs. long category boundary perception in Estonian. Since the intrinsic duration of a close vowel /i/ is about 10-15 ms shorter than that of an open vowel /a/, we hypothesize that the short vs. long category boundary in /i/ occurs at a shorter duration than in the case of...
متن کامل