Causal Inference

Two current projects have causal inference goals.

Causal Inference in Observed Populations

NoteWith Tim Johnson

We have a long-standing project expanding on his work using the Central Personnel Data File from the United State Office of Personnel Management. In this paper, we examine the interaction of veteran's preference and education to demonstrate that the policy has no observable effect in the population. The computation for this problem has proven a significant challenge. The paper received a revise and resubmit at Statistics and Public Policy years ago but we knew of a problem that we only recently fixed.

Status: Under revision for resubmission
Code: GitHub Repository

Smoking Project

With Debra Ringold and Kawika Pierson, we examine the role of deterministic and probabilistic language in individual risk perceptions of smoking related health conditions. We thank the Center for Governance and Public Policy and it’s director Tim Johnson for funding assistance with the project.

General Interest

Judea Pearl’s Causality and The Book of Why provide a powerful foundation for the investigation of causal relationships. For artificial intelligence, which I am keenly interested in, the ladder of causality in Pearl is of profound interest.