mwa

Sebastian Schutte and Karsten Donnay developed a novel method for causal inference in spatiotemporal event data called Matched Wake Analysis (mwa). The corresponding R package computationally efficiently implements the methodology in a user-friendly interface. The package is designed to analyze causal relationships in spatially and temporally referenced data. Specific types of events might affect subsequent levels of other events. To estimate the corresponding effect, treatment, control, and dependent events are selected from the empirical sample. Treatment effects are established through automated matching and a diff-in-diffs regression design. The analysis is repeated for various spatial and temporal offsets from the treatment events.

The package is available on CRAN  and on  GitHub, for details on the method please refer to Schutte & Donnay (2014).