Marianthi Markatou


Marianthi Markatou


Research Topics

Statistical Sciences (Statistics and Biostatistics): robustness (studies the sensitivity of inferential procedures to deviations from the hypothesized assumptions), mixture models (statistical models for accommodating data heterogeneity), statistical distances, high dimensional data analysis, development of inference methods for big data, goodness of fit problems, problems in model assessment and selection, classification and clustering, surveillance in large databases, machine learning and kernel methods, biomarker development and ROC analysis, analysis of computer algorithms, studies on dependence (theory of copulas and general dependence measures) Interdisciplinary: biomedical informatics, text data mining, emerging safety sciences relevant to health, study of dependence in microarrays and proteomics data, methodology development for comparative effectiveness research, methods development for comparative safety research, data mining methods and applications in biomedicine and public health, statistical learning methods for drug repurposing

Contact Information

726 Kimball Tower

Buffalo NY, 14214

Phone: (716) 829-2894

Fax: (716) 829-2200

Education and Training

  • Certificate of Training, Biomedical Informatics, Columbia University
  • PhD, Statistics, The Pennsylvania State University
  • MA, Statistics, University of Rochester
  • BS, Mathematics/Physics, University of Patras, Greece

Professional Affiliations

  • American Statistical Association, Elected Fellow
  • BMRD Study Section (2014-2020), Permanent Member
  • Institute of Mathematical Statistics
  • International Biometric Society
  • International Statistical Institute, Elected Member
  • Scientific Advisory Committee Member, IMEDS, Reagan-Udall Foundation for the FDA


  • PCORI: Patient-centered HCV care via Telemedicine for Individuals on Opiate Substitutions Therapy: A stepped-wedge Clustered Randomized Controlled Trial 
  • NIH/NIDA: Pharmacokinetics of Methadone in HIV/HCV co-infected patients with and without liver fibrosis.