Learning by Convex Combination

Karol Flores-Szwagrzak

No 16-2022, Working Papers from Copenhagen Business School, Department of Economics

Abstract: We study how an agent evaluates the optimality of an action when she only observes a sample of its outcomes, not the outcome distribution. We characterize a model where the agent assigns an ex-ante utility to the action and then, upon seeing the sample, “updates” her evaluation by taking a convex combination of this ex-ante utility and the average utility of the outcomes in the sample. The weight on the average utility in this convex combination increases with sample size. Asymptotically, actions are evaluated using their sample average utility. The model includes Bayesian benchmarks as special cases. More generally, it describes an agent that may learn imperfectly yet consistently with a rough intuitive understanding of the Law of Large Numbers; it also enables decision-theoretic definitions of important concepts in the descriptive study of probabilistic judgement.

Keywords: Sample; Sample size; Learning; uncertainty (search for similar items in EconPapers)
JEL-codes: D81 D83 (search for similar items in EconPapers)
Pages: 51 pages
Date: 2022-11-24
New Economics Papers: this item is included in nep-mic and nep-upt
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://hdl.handle.net/10398/cccde635-5980-4348-ac2e-9ea6acf6e3d5 Full text (application/pdf)
Full text not avaiable

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hhs:cbsnow:2022_016

Ordering information: This working paper can be ordered from

Access Statistics for this paper

More papers in Working Papers from Copenhagen Business School, Department of Economics Copenhagen Business School, Department of Economics, Porcelaenshaven 16 A. 1.floor, DK-2000 Frederiksberg, Denmark. Contact information at EDIRC.
Bibliographic data for series maintained by CBS Library Research Registration Team ().