Spillovers of Program Benefits with Missing Network Links

21 Sep 2020  ·  Lina Zhang ·

The issue of missing network links in partially observed networks is frequently neglected in empirical studies. This paper aims to address this issue when investigating spillovers of program benefits in the presence of network interactions. We focus on a nonparametric outcome model that allows for flexible forms of heterogeneity, and we propose two methods to point identify the treatment and spillover effects in the case of bounded network degree. When the network degree is unbounded, these methods can be used to mitigate the bias caused by missing network links. The first method utilizes two easily-constructed network measures based on the incoming and outgoing links. The second method relies on a single network measure with a known rate of missing links. We demonstrate the effective bias reduction provided by our methods using Monte Carlo experiments and a naturalistic simulation on real-world school network data. Finally, we re-examine the spillovers of home computer use on children's self-empowered learning.

PDF Abstract
No code implementations yet. Submit your code now

Tasks


Datasets


  Add Datasets introduced or used in this paper

Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods


No methods listed for this paper. Add relevant methods here