When Does Learning in Games Generate Convergence to Nash Equilibria? The Role of Supermodularity in an Experimental Setting

Yan Chen and Robert S. Gazzale
Additional contact information
Yan Chen: University of Michigan
Robert S. Gazzale: Williams College

No 2004-02, Department of Economics Working Papers from Department of Economics, Williams College

Abstract: This study clarifies the conditions under which learning in games produces convergence to Nash equilibria in practice. Previous work has identified theoretical conditions under which various stylized learning processes achieve convergence. One technical condition is supermodularity, which is closely related to the more familiar concept of strategic complementarities. We experimentally investigate the role of supermodularity in achieving convergence through learning. Using a game from the literature on solutions to externalities, we systematically vary a free parameter below, close to, at and beyond the threshold of supermodularity to assess its effects on convergence. We find that supermodular and "near-supermodular" games converge significantly better than those far below the threshold. From a little below the threshold to the threshold, the improvement is statistically insignificant. Within the class of supermodular games, increasing the parameter far beyond the threshold does not significantly improve convergence. Simulation shows that while most experimental results persist in the long run, some become more pronounced.

Keywords: learning; supermodular games (search for similar items in EconPapers)
JEL-codes: C90 D70 (search for similar items in EconPapers)
Pages: 56 pages
Date: 2004-02
References: Add references at CitEc
Citations:

Published in The American Economic Review, 94(5), December 2004, 1505-1503.

Downloads: (external link)
http://lanfiles.williams.edu/~rgazzale/research/comp20031022.pdf
Our link check indicates that this URL is bad, the error code is: 500 Can't connect to lanfiles.williams.edu:80 (nodename nor servname provided, or not known)

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:wil:wileco:2004-02

Ordering information: This working paper can be ordered from

Access Statistics for this paper

More papers in Department of Economics Working Papers from Department of Economics, Williams College Williamstown, MA 01267. Contact information at EDIRC.
Bibliographic data for series maintained by Greg Phelan ().