Measuring external stability in one-to-one matching

Abstract

We consider a one-to-one matching model where the population expands with the arrival of a new individual. Individuals in this population are matched according to the deferred acceptance (DA) algorithm where men propose and women accept or reject. Using computer simulations of this model, we study how the matching utilities of individuals are affected when the initial size of the population or the degree of heterogeneity in preferences change. We also investigate to what extent the preference heterogeneity affects the difference between the likelihoods of matching obtained under the DA algorithm and the Top Trading Cycle algorithm.