I am a Research Scientist at Ataraxis AI, where I build cutting-edge AI models for cancer prognosis and treatment selection.
Before joining Ataraxis AI, I was a Postdoctoral Fellow in the Division of Biostatistics at NYU Grossman School of Medicine, where I worked with Iván Díaz and Wenbo Wu on theory and methods at the intersection of statistics, machine learning, and causal inference, with applications in healthcare. I also collaborated with the Tech & Society Lab to assess the causal effects of social media use on mental health. I received my PhD in Statistics from Carnegie Mellon University, where I was advised by Zach Branson and Edward Kennedy, and worked closely with Siva Balakrishnan and Larry Wasserman.
My research develops nonparametric statistical methods that leverage machine learning to estimate causal effects from complex observational and longitudinal data, with an emphasis on robustness to assumption violations and statistical efficiency in high-dimensional settings. This work has been recognized with the Tom Ten Have Award for exceptional research in causal inference. Recently, I have focused on longitudinal causal inference, developing new interventions and efficient sequential double machine learning estimators that address positivity violations.
Google Scholar and CV (current as of October 2025)
Non-overlap average treatment effect bounds
H. Susmann*, A.McClean*, and I. Díaz
arxiv, bluesky
* Equal contribution
Propensity score weighting across counterfactual worlds: longitudinal effects under positivity violations
A. McClean and I. Díaz
arxiv, bluesky
Longitudinal weighted and trimmed treatment effects with flip interventions
A. McClean, A. Levis, N. Williams, and I. Díaz
arxiv, bluesky, slides
Fair comparisons of causal parameters with many treatments and positivity violations
A. McClean, Y. Li, S. Bae, M. A. McAdams-DeMarco, I. Díaz, and W. Wu
R and R at Biometrika
arxiv, bluesky, slides
Stochastic interventions, sensitivity analysis, and optimal transport
A. Levis, E. H. Kennedy, A. McClean, S. Balakrishnan, and L. Wasserman
arxiv
Calibrated sensitivity models
A. McClean, Z. Branson, and E.H. Kennedy
Minor revision at Biometrika
arxiv, bluesky, slides
Double Cross-fit Doubly Robust Estimators: Beyond Series Regression
A. McClean, S. Balakrishnan, E.H. Kennedy, and L. Wasserman
Winner of Tom Ten Have award at ACIC 2023, Major revision at JRSSB
arxiv, bluesky
Nonparametric Estimation of Conditional Incremental Effects
A. McClean, Z. Branson, and E. H. Kennedy
Journal of Causal Inference, 2024, journal, arxiv
Incremental causal effects: an introduction and review
M. Bonvini*, A. McClean*, Z. Branson, and E. H. Kennedy
Handbook of Matching and Weighting in Causal Inference, 2023, arxiv
* Equal contribution
Incremental Propensity Score Effects for Criminology: An Application Assessing the Relationship Between Homelessness, Behavioral Health Problems, and Recidivism
L. Jacobs, A. McClean, Z. Branson, E. H. Kennedy, and A. Fixler
Journal of Quantitative Criminology, 2023, journal, arxiv
Contributor to npcausal package
(Sept ‘25) Our new paper on non-overlap average treatment effect bounds is on arxiv
(Jul ‘25) Our new paper on counterfactual longitudinal propensity score weighting is on arxiv
(June ‘25) Our new paper on flip interventions for weighting and trimmed with longitudinal data is on arxiv
(Oct ‘24) We have a new paper on arxiv on fair comparisons of causal parameters; more details on Bluesky
(May ‘24) I presented our ongoing work on calibrated sensitivity models at ACIC 2024 (slides). A working paper is up on arxiv
(Mar ‘24) We have a new paper on arxiv on double cross-fit doubly robust estimators!