Presenter: Jason Saporta, PhD Candidate, Department of Statistics, Iowa State University

Title: Statistical Tools for Causal Inference and Forensic Science

Abstract: I present two manuscripts on the problem of partial identification in causal inference, and a third on an application of machine learning in bloodstain pattern analysis. Partial identification concerns the bounding of causal quantities that remain unidentifiable given the observed distribution and the causal diagram of the underlying structural causal model (SCM). I present an extension to current methods for Bayesian partial identification, further increasing the range of models it can be applied to. This is followed by a description of a software library designed to make partial identification easier to apply in practice. The final manuscript presents an automated method for identifying the point of origin of a given bloodstain pattern. I conclude with a brief description of recent research that may help causal inference techniques see more widespread adoption in forensic science problems.