Link Search Menu Expand Document

Welcome to the AME Lab!

The Almost Matching Exactly Lab provides a range of matching methods for causal inference using statistical machine learning algorithms.

View us on GitHub


The Almost Matching Exactly Lab is a joint venture of the Departments of Computer Science and Statistics at Duke University in Durham, North Carolina. Our goal is to develop and apply interpretable machine learning algorithms to estimate causal effects using observational data. In general, our algorithms match units with similar covariate distributions, creating high quality, exact or almost exact matches for treatment effect estimation. To learn more about how these algorithms work, visit our algorithm overview page or read one of our publications. To begin using one of our matching methods, choose a software package and get started!