Faculty Profile

Michele Peruzzi

Michele Peruzzi, PhD

Assistant Professor, Biostatistics
I am interested in the development of Bayesian methods for multivariate correlated data, with a specific focus on geostatistics for environmental health applications as well as on imaging and high resolution sensor data. In addition to methods development, I am active in building new software packages for reproducible and scalable statistical analyses of large scale datasets.

  • PhD, Statistics, Università Bocconi, 2019
  • MSc, Economic and Social Sciences, Università Bocconi, 2013
  • BSc, Economic and Social Sciences, Università Bocconi, 2010

My research interests include spatial statistics, Bayesian methods, models for complex sensor data, environmental health, and climate change.

My current projects involve the development of scalable Bayesian methods for (1) analyzing complex, correlated, multivariate high resolution sensor data, including satellite imaging, air quality sensors, and medical imaging; (2) characterizing complex interaction effects in response surface models for large scale health data; (3) understanding the impacts of climate change on natural ecosystems and human health.

M. Peruzzi, S. Banerjee & A.O. Finley (2022), Highly Scalable Bayesian Geostatistical Modeling via Meshed Gaussian Processes on Partitoned Domains. Journal of the American Statistical Association 117(538): 969-982. https://www.tandfonline.com/doi/full/10.1080/01621459.2020.1833889

M. Peruzzi & D.B. Dunson (2022), Spatial Multivariate Trees for Big Data Bayesian Regression. Journal of Machine Learning Research 23(17):1−40. https://www.jmlr.org/papers/v23/20-1361.html

N. Neupane, M. Peruzzi, L. Ries, A. Arab, S. J. Mayor, J. C. Withey, A. O. Finley (2022) A novel model to accurately predict continental-scale green-up timing. International J. of Applied Earth Observation and Geoinformation 108:102747 https://doi.org/10.1016/j.jag.2022.102747

M4531 SPH II
1415 Washington Heights
Ann Arbor, MI 48109-2029

Email: [email protected]

Areas of Expertise: Biostatistics