Mathematics & Statistics Colloquium
Causal Inference Powered by Satellite Data: Methods for Informing Environmental Policy
Megan Ayers (‘19)
Assistant Professor of Statistics
Reed College
What interventions and policies most effectively mitigate climate change and protect the environment? Causal inference methods can provide relevant scientific evidence for this question by rigorously demonstrating cause-and-effect relationships. However, causal assumptions often break down in these complex, real-world contexts. Satellite data coupled with machine learning methods offers an inexpensive, accessible, and detailed data source for analyses, but systematic measurement errors can introduce bias into causal estimates. This talk will focus on work that addresses challenges in evaluation of incentive-based forest conservation programs. These topics illustrate the potential of combining statistics, machine learning, and satellite data to address pressing environmental challenges.
March 19, 2026, JR Howard 132, 4:00-5:00
All are welcome!