Aggregate Matchings
- Creators
-
Echenique, Federico
- Lee, Sangmok
-
Shum, Matthew
Abstract
This paper characterizes the testable implications of stability for aggregate matchings. We consider data on matchings where individuals are aggregated, based on their observable characteristics, into types, and we know how many agents of each type match. We derive stability conditions for an aggregate matching, and, based on these, provide a simple necessary and sufficient condition for an observed aggregate matching to be rationalizable (i.e. such that preferences can be found so that the observed aggregate matching is stable). Subsequently, we derive moment inequalities based on the stability conditions, and provide an empirical illustration using the cross-sectional marriage distributions across the US states.
Additional Information
Echenique thanks the Lee Center at Caltech, and the NSF, for funding related to this project. We also thank Pierre-Andre Chiappori, Aureo de Paula, Bernard Salanie, Leeat Yariv, and seminar participants at Boston College, Universitat Aut`onoma de Barcelona, Universidad de Vigo, UCLA, the 2010 CAPCP conference at Penn State, the Roth-Sotomayor Celebration in Duke University, and the 2010 Cowles Summer Conference at Yale, for helpful comments. Supplemental material includes 99 slides for the NBER Market Design Conference (Oct. 8, 2010)Attached Files
Accepted Version - sswp1316.pdf
Files
| Name | Size | Download all |
|---|---|---|
|
md5:a0472d5f1cffffb1d1cbd7c1206054a0
|
449.1 kB | Preview Download |
Additional details
- Eprint ID
- 20351
- Resolver ID
- CaltechAUTHORS:20101008-101034676
- Caltech Lee Center for Advanced Networking
- NSF
- Created
-
2010-10-08Created from EPrint's datestamp field
- Updated
-
2019-11-26Created from EPrint's last_modified field
- Caltech groups
- Social Science Working Papers