Professor Michael Schweinberger, Pennsylvania State University
A hybrid event organised by the Department of Statistics and Cramérsällskapet.
Regression under interference in connected populations: models, methods, and theory
Abstract
To understand how the interconnected and interdependent world of the twenty-first century operates and make model-based predictions, joint probability models for networks and interdependent outcomes are needed. We propose a comprehensive regression framework for networks and interdependent outcomes with multiple advantages, including interpretability, scalability, and provable theoretical guarantees. The regression framework can be used for studying relationships among attributes of connected units and captures complex dependencies among connections and attributes, while retaining the virtues of linear regression, logistic regression, and other regression models by being interpretable and widely applicable. On the computational side, we show that the regression framework is amenable to scalable statistical computing based on convex optimization of pseudo-likelihoods using minorization-maximization methods. On the theoretical side, we introduce a novel approach to obtaining rates of convergence for M-estimators in exponential families based on the idea of "transporting" concentration of measure results between homeomorphic spaces, which facilitates rates of convergence for pseudo-likelihood estimators based on a single observation of dependent connections and attributes. The dependence among connections and attributes is controlled using coupling methods and percolation theory. We demonstrate the regression framework using simulations and an application to hate speech on the social media platform X.
Senast uppdaterad: 2026-05-05
Sidansvarig: Statistiska institutionen