18 août 2020
We introduce an approach to deal with self-selection of peers in the linear-in-means model. Contrary to the existing proposals we do not require to specify a model for how the selection of peers comes about. Rather, we exploit two restrictions that are inherent to many such specifications to construct intuitive instrumental variables. These restrictions are that link decisions that involve a given individual are not all independent of one another, but that they are independent of the link behavior between other pairs of individuals. A two-stage least-squares estimator of the linear-in-means model is then readily obtained.