Research Repository

Second-Order Induction in Prediction Problems

Argenziano, Rossella and Gilboa, Itzhak (2019) 'Second-Order Induction in Prediction Problems.' Proceedings of the National Academy of Sciences, 116 (21). pp. 10323-10328. ISSN 0027-8424

argenziano_gilboa_2nd_order_induction.pdf - Accepted Version

Download (367kB) | Preview
Second_Order_Induction_SI_2019_03_21.pdf - Supplemental Material

Download (267kB) | Preview


Agents make predictions based on similar past cases, while also learning the relative importance of various attributes in judging similarity. We ask whether the resulting "empirically optimal similarity function” (EOSF) is unique, and how easy it is to find it. We show that with many observations and few relevant variables, uniqueness holds. By contrast, when there are many variables relative to observations, non-uniqueness is the rule, and finding the EOSF is computationally hard. The results are interpreted as providing conditions under which rational agents who have access to the same observations are likely to converge on the same predictions, and conditions under which they may entertain different probabilistic beliefs.

Item Type: Article
Uncontrolled Keywords: Learning; Empirically Optimal Similarity Function; Belief Formation; Kernel Estimation; Generalized Context Model
Subjects: H Social Sciences > HB Economic Theory
Divisions: Faculty of Social Sciences
Faculty of Social Sciences > Economics, Department of
SWORD Depositor: Elements
Depositing User: Elements
Date Deposited: 08 May 2019 08:41
Last Modified: 06 Jan 2022 13:59

Actions (login required)

View Item View Item