Debashis Ghosh, Ph.D.

Debashis Ghosh, Ph.D.

Professor, Department of Statistics and Public Health Sciences

Investigator, The Methodology Center


The Pennsylvania State University

514A Wartik Lab

University Park, PA 16802







Ph.D., University of Washington, 2000 (Biostatistics)

M.S., University of Washington, 1997 (Applied Statistics)

B.A., Rice University 1995 (Mathematics, Statistics, Economics and French)




Current Projects and Collaborations

I work with Donna Coffman of The Methodology Center on the development and application of new statistical methodology for inferring causal effects with observational data. In particular, I am interested in the development of new machine learning methodologies for the analysis of observational data.



Research Interests

  • Causal inference
  • Machine learning
  • Observational data

Honors and Awards

  • 2011: Faculty Scholar, Methodology Center, Pennsylvania State University
  • 2010-2012: College of CSR Reviewers, National Institutes of Health
  • 2008: Featured Biomed Central Editorial Board Member
  • 2008: Biometrics Best Paper Award
  • 2006: UM CCMB Pilot Grant Award




Statistical Methods for Cancer Biomarkers

NIH R01CA129102

This grant proposes new semiparametric and nonparametric modelling procedures with applications to cancer biomarkers. The aims are the following: (a) Development of semiparametric and nonparametric multivariate isotonic regression modelling procedures for biomarkers; (b) Development of statistical methods for the analysis of surrogate endpoints in a single-trial and multiple-trial framework; (c) Development of hybrid model averaging methods and attendant projection-based framework for combining biomarkers to optimize predictive accuracy.

1/1/2009-present; Role: Co-Principal Investigator (PI of Penn State subcontract; PI: J. M. G. Taylor)


Penn State Clinical and Translational Institute

NIH UL1RR033184

This proposal describes the newly created Penn State Clinical and Translational Science Institute (CTSI). The overarching goal of our CTSI initiative is to revitalize the health science research and education enterprise at our University to better enable it to deliver on the promise of improved health.

09/01/11 - present; Role: Biostatistician (PI: L. Sinoway)



Selected Recent Publications

Ghosh, D., Taylor, J. M., & Sargent, D. J. (2012). Meta-analysis for surrogacy: Accelerated failure time modelling and semi-competing risks (with discussion). Biometrics, 68(1), 226-232. PMCID: PMC3175271

Ghosh, D. (2011). Propensity score modelling in observational studies using dimension reduction methods. Statistics & Probability Letters, 81(7), 813-820. PMCID: PMC3099445



Coffman, D. L., MacKinnon, D. P., Zhu, Y., & Ghosh, D. (2013, January). Comparison of three approaches to causal mediation analysis. Presented at the Causal Mediation Analysis Symposium, Ghent, Belgium.

Coffman, D. L., MacKinnon, D. P., Zhu, Y., & Ghosh, D. (2012, June). Comparison of three approaches to causal mediation analysis. In Thoemmes, F. (Chair), Causal inference in mediation models. Poster presented at the 20th Annual Meeting of the Society for Prevention Research, Washington, DC.

Zhu, Y., Coffman, D. L., MacKinnon, D. P., & Ghosh, D. (2012, May). Comparison of three approaches to causal mediation analysis. Poster presented at the 2012 Atlantic Causal Inference Conference, Baltimore, MD.

Follow Us