Dynamic Directed Random Matching
We develop a general and unified model in which a continuum of agents conduct directed random searches for counterparties. Our results provide the first probabilistic foundation for static and dynamic models of directed search (including the matching-function approach) that are common in search-based models of financial markets, monetary theory, and labor economics. The agents' types are shown to be independent discrete-time Markov processes that incorporate the effects of random mutation, random matching with match-induced type changes, and with the potential for enduring partnerships that may have randomly timed break-ups. The multi-period cross-sectional distribution of types is shown to be deterministic and is calculated using the exact law of large numbers.
Part of this work was presented at the Asian Meeting of the Econometric Society in Singapore in August 2013 and in Taipei in June 2014; at the PIMS Summer School on the Economics and Mathematics of Systemic Risk and the Financial Networks in the Pacific Institute for the Mathematical Sciences, Vancouver, July 2014; and at the World Congress of the Econometric Society, Montreal, August 2015. We are grateful for comments from Peter Loeb. This version owes substantially to the careful reading and expository suggestions of an editor, an associate editor and the referees. The work was partially supported by the NUS grants R-122-000-227-112 and R146-000-215-112. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
Darrell Duffie & Lei Qiao & Yeneng Sun, 2018. "Dynamic directed random matching," Journal of Economic Theory, vol 174, pages 124-183. citation courtesy of