About me

Constanza finished her PhD in Social Statistics at the University of Manchester, specializing in causal inference methods for evaluating health policies. Her doctoral work develops novel doubly robust machine learning estimators—combining experimental and observational data—to generalize randomized trial findings to national target populations in non-nested designs.

This research positions her at the intersection of causal modeling, food labeling policy, and population-level inference. Prior to her doctorate, she built and directed national data infrastructure at Chile’s National Institute of Statistics, providing rare expertise in large-scale food system datasets, multistage survey design, and algorithmic data quality control at census scale.

Research Interests

• Causal Inference: Transportability & generalization of trial findings
• Machine Learning: Doubly robust methods (BART, Gradient Boosting)
• Food Systems: Modeling and evaluating health policies
• Survey Design: Complex multistage designs and population-level inference