On Statistical Discrimination as a Failure of Social Learning: A Multiarmed Bandit Approach
この論文をさがす
説明
<jats:p> We analyze statistical discrimination in hiring markets using a multiarmed bandit model. Myopic firms face workers arriving with heterogeneous observable characteristics. The association between the worker’s skill and characteristics is unknown ex ante; thus, firms need to learn it. Laissez-faire causes perpetual underestimation: minority workers are rarely hired, and therefore, the underestimation tends to persist. Even a marginal imbalance in the population ratio frequently results in perpetual underestimation. We demonstrate that a subsidy rule that is implemented as temporary affirmative action effectively alleviates discrimination stemming from insufficient data. </jats:p><jats:p> This paper was accepted by Nicolas Stier-Moses, Management Science Special Issue on The Human-Algorithm Connection. </jats:p><jats:p> Funding: This work was supported by the Social Sciences and Humanities Research Council of Canada [Grant 430-2020-00088] and JST ERATO [Grant JPMJER2301], Japan. </jats:p><jats:p> Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.00893 . </jats:p>
収録刊行物
-
- Management Science
-
Management Science 2024-03-29
Institute for Operations Research and the Management Sciences (INFORMS)
- Tweet
キーワード
- FOS: Computer and information sciences
- Econometrics (econ.EM)
- Machine Learning (stat.ML)
- FOS: Economics and business
- Computer Science - Computer Science and Game Theory
- Statistics - Machine Learning
- Economics - Theoretical Economics
- Theoretical Economics (econ.TH)
- Economics - Econometrics
- Computer Science and Game Theory (cs.GT)
詳細情報 詳細情報について
-
- CRID
- 1870302167531482880
-
- ISSN
- 15265501
- 00251909
-
- データソース種別
-
- OpenAIRE