My work is devoted to finding new, rigorous ways to make causal inference in the absence of experiments and without using parametric assumptions.
My ultimate goal is to find better ways to estimate the effect policies may have on people, in order to inform better decision-making. I am interested broadly in policy problems around inequity, and have worked on questions in the criminal justice system, hiring, and financial services.
As a graduate student, I worked with Edward Kennedy to develop doubly robust causal estimators and for social science applications. In collaboration with Daniel Nagin, I applied these estimators to questions about the criminal justice system.
Works in Progress:
Mauro, J., Kennedy, E., Nagin, D. Instrumental Variables Methods Using Dynamic Interventions. http://arxiv.org/abs/1811.01301 (Revise and Resubmit at JRSS-A)
Mauro, J., Blumenstock, J., Yen, K., Linxie, A. Nonparametric Causal Estimators for Multivariate Missing Data: An Application to Estimate Treatment Effects from Digital Trace Data
Mauro, J., Kennedy, E., Jacquillat, A., Nagin, D. Causal Estimation of Resource Constrained Optimal Treatments.
Cuellar, M., Mauro, J., When Additional Information Increases Error: Withholding biasing information from Forensic Scientists