Essays in Empirical Matching

Abstract

This thesis combines three essays on empirical applications and methods in two-sided matching markets. The first essay uses existing methods to estimate preferences for schools using rank order lists from New York City's new high school assignment system launched in Fall 2003 to study the consequences of coordinating school admissions in a mechanism based on the student-proposing deferred acceptance algorithm. The second essay develops techniques for estimating preferences in two-sided matching markets with non-transferable utility using only data on final matches. It uses these techniques to estimate preferences in the market for family medicine residents. These estimates are then used to analyze two economic questions. First, it investigates whether centralization in the market for medical residents is primarily responsible for low salaries paid to medical residents. Second, it analyzes the effects of government interventions intended to encourage training of medical residents in rural areas. The final essay studies estimation and non-parametric identfication of preferences in two-sided matching markets with non-transferable utility. It studies the special case in which preferences of each side of the market is vertical and data from a pairwise stable match, in a single large market is observed.