Project Grant 2413954

This $200,000 federal Project Grant award from the National Science Foundation (NSF) Division of Mathematical Sciences (CFDA 47.049 - Mathematical and Physical Sciences) supports research on the design and analysis of experiments under the K-Nearest-Neighbors Interference Model (KNNIM). The project will derive estimators for causal estimands like Nearest Neighbor Treatment Effects, develop tests to determine if relaxed KNNIM models are plausible, and derive effective experimental designs for improved estimation and inference of treatment effects under KNNIM. Additionally, the project will create methods to simultaneously determine the correct KNNIM interference structure and estimate treatment effects through repeated experimentation. The research aims to advance scientific knowledge and techniques in areas where traditional causal inference methods fail to account for interference effects, such as in social media network experiments. The award also provides research training opportunities for graduate students at Kansas State University, the primary recipient of this grant.