This research project will advance methods of estimation and inference for large-scale network models. The modeling of network data has important applications in many areas, including sociology, economics, marketing, and public health. The project will extend methods from the one-mode network setting (such as social media networks) to the bipartite network setting (such as worker-employer networks). Using data from the U.S. Census Bureau and the newly developed methods, the project will examine the impact of immigration on the U.S. job market over time. Theoretical and empirical analyses will be complemented by the development and dissemination of open source software that implements the new methods and enables analyses of large-scale social and economic networks. The project's interdisciplinary team of researchers will train and mentor graduate students in the social sciences and statistics.

This research project will develop new modeling capabilities and statistical inference for latent space network models, specifically generalized random dot product graphs. An overarching goal of the project will be to disentangle the effect of observable and unobservable node characteristics on the process of link formation in a manner that is both statistically principled and computationally feasible. Project results will include methods for both one-mode networks and bipartite networks, as well as for both static and dynamic networks. Asymptotic spectral theory and computationally efficient procedures for understanding bipartite networks will be developed. Spectral estimates for bipartite networks will be used to correct for network endogeneity in regression models. Combining spectral approaches with classical regression procedures from network econometrics will facilitate the development of effective alternatives to existing methods in the analysis of how networks impact socioeconomic outcomes. Using these new methods, the project will examine the impact of immigration on the U.S. job market with data from the Longitudinal Employer-Household Dynamics program.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

National Science Foundation (NSF)
Division of Social and Economic Sciences (SES)
Application #
Program Officer
Cheryl Eavey
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Johns Hopkins University
United States
Zip Code