Estimation of Games under No Regret: Structural Econometrics for AI

Niccolò Lomys and Lorenzo Magnolfi ()
Additional contact information
Lorenzo Magnolfi: Department of Economics, University of Wisconsin-Madison

No 24-05, Working Papers from NET Institute

Abstract: We develop a method to recover primitives from data generated by artificial intelligence (AI) agents in strategic environments such as online marketplaces and auctions. Building on how leading online learning AIs are designed, we assume agents minimize their regret. Under asymptotic no regret, we show that time-average play converges to the set of Bayes coarse correlated equilibrium (BCCE) predictions. Our econometric procedure is based on BCCE restrictions and convergence rates of regret-minimizing AIs. We apply the method to pricing data in a digital marketplace for used smartphones. We estimate sellers' cost distributions and find lower markups than in centralized platforms.

Keywords: AI Decision-Making; Empirical Games; Regret Minimization; Bayes (Coarse) Correlated Equilibrium; Partial Identification (search for similar items in EconPapers)
JEL-codes: C1 C5 C7 D4 D8 L1 L8 (search for similar items in EconPapers)
Pages: 41 pages
Date: 2024-09, Revised 2024-11
New Economics Papers: this item is included in nep-ain and nep-ecm
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.netinst.org/Lomys_24-05.pdf (application/pdf)
no

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:net:wpaper:2405

Access Statistics for this paper

More papers in Working Papers from NET Institute
Bibliographic data for series maintained by Nicholas Economides ().