Identify experts through Revealed Confidence : application to Wisdom of Crowds
Name
1227097094-MIT.pdf
Size
910.85 KB
Format
Adobe PDF
Checksum (MD5)
21baea6b18b712e4f3a390fd312e9a01
Author(s)
Zhang, Yunhao(Business management scientist)Massachusetts Institute of Technology.
Advisor(s)
Drazen Prelec.
Alternative Title
Identify experts through RC : application to WoC
Date Issued
2020
Publisher
Massachusetts Institute of Technology
Abstract
We propose our Revealed Confidence (RC) algorithm that improves Wisdom of Crowds (WoC) by identifying experts from the crowds. We highlight the important distinction between first- and second-order uncertainty, which also serves as an explanation for rational overconfidence. Under our proposed belief updating mechanism, we analyze the performance of RC algorithm and show the algorithm could identify the more accurate prior estimates even if all agents report the same prior confidence under conventional confidence elicitation, e.g. confidence interval. Our empirical analysis shows that (1) RC improves upon other wisdom of Crowds methods by overweighting the more accurate agents in the aggregation (2) verifies one key prediction of our theoretical result that the distance effect indeed affects belief-updating henceforth RC algorithm's performance, which should be carefully controlled for in order to optimize the algorithm..
Description
Thesis: S.M. in Management Research, Massachusetts Institute of Technology, Sloan School of Management, September, 2020
Cataloged from student-submitted PDF version of thesis.
Includes bibliographical references (pages 52-54).
Subjects
Sloan School of Management.
MIT Department
Sloan School of Management
Terms of Use
MIT theses may be protected by copyright. Please reuse MIT thesis content according to the MIT Libraries Permissions Policy, which is available through the URL provided.
Persistent DSpace Link
