Learning Source Biases: Multisource Misspecifications and Their Impact on Predictions

50 Pages Posted: 4 Oct 2024

See all articles by Junnan He

Junnan He

Sciences Po

Lin Hu

Australian National University (ANU)

Matthew Kovach

Mitchell E. Daniels, Jr School of Business, Purdue University

Anqi Li

University of Waterloo - Department of Economics

Date Written: September 01, 2024

Abstract

We study how a Bayesian decision maker (DM) learns about the biases of novel information sources to predict a random state. Absent frictions, the DM uses familiar sources as yardsticks to accurately discern the biases of novel sources. We derive the distortion of the DM's long-run prediction when the DM holds misspecified beliefs about the biases of several familiar sources. The distortion aggregates misspecifications across familiar sources independently of the number and nature of the novel sources the DM learns about. This has implications for labor market discrimination, media bias, and project finance and oversight.

Keywords: Misspecified Learning, Discrimination, Beliefs

Suggested Citation

He, Junnan and Hu, Lin and Kovach, Matthew and Li, Anqi, Learning Source Biases: Multisource Misspecifications and Their Impact on Predictions (September 01, 2024). Available at SSRN: https://ssrn.com/abstract=4943579 or http://dx.doi.org/10.2139/ssrn.4943579

Junnan He

Sciences Po ( email )

Lin Hu

Australian National University (ANU) ( email )

Canberra, Australian Capital Territory 2601
Australia

Matthew Kovach (Contact Author)

Mitchell E. Daniels, Jr School of Business, Purdue University ( email )

403 Mitch Daniels Blvd.
West Lafayette, IN 47907
United States

Anqi Li

University of Waterloo - Department of Economics ( email )

Waterloo, Ontario

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
37
Abstract Views
139
PlumX Metrics