Assistant University
Resume:
SURAJIT RAY
Department of Mathematics and Statistics
B: ****@****.**.***
Boston University, Boston, MA 02215
Homepage: http://math.bu.edu/people/sray/
Professional Experience
2006 Assistant Professor, Dept. of Mathematics and Statistics,
Boston University, Boston.
2005 2006 Visiting Assistant Professor, Dept. of Biostatistics,
University of North Carolina at Chapel Hill., Chapel Hill
2004 2005 Post Doctoral Fellow, Statistics and Applied Mathematical Sciences Institute,
Research Triangle Park, Durham
2003 2004 Visiting Assistant Professor, Dept. of Biostatistics
University of North Carolina at Chapel Hill.
2000 2003 Research Assistant, Dept. of Statistics,
Pennsylvania State University, University Park
Education
Aug 2003 Ph.D. Dept of Statistics, Pennsylvania State University
Dissertation: Distance-based Model-Selection with application to Analysis of Gene
Expression Data
Advisor: Bruce G. Lindsay.
Dates attended: Aug 1999- Aug 2003
1999 M. Stat., Indian Statistical Institute, Calcutta, India.
First Division with distinction;
Specialization: Applied Statistics and Data Analysis.
Dates attended: Aug 1997- May 1999
1997 B. Sc. (Honors) in Statistics, Presidency College, Calcutta
First Division with distinction;
Minors: Mathematics, Economics.
Dates attended: Aug 1994- May 1997
Sep, 2008 1 of 7
Honors and Awards
2007 Honored by the Class of 2007 through the Class of 2007 Gift Program at Boston
University.
2004 Laha Travel Award from the Institute of Mathematical Statistics
2003 Most Outstanding Student Presentation in Theoretical Statistics at the Interna-
tional Conference on Statistics, Combinatorics and Related Areas. (See Presenta-
tions below)
2001-2003 Several graduate student travel awards from the Eberly College of Science,
PennState.
2002 Davey Graduate fellowship award from the Eberly College of Science, PennState.
2002 August and Ruth Homeyer Graduate fellowship award from the Eberly College of
Science, PennState.
2000 Vollmer-Kleckner Scholarship award in Science from the Eberly College of Science,
PennState, for the most outstanding performance in PhD Quali ers.
Research Interests
Theory and applications of nite mixture models, and detection of modes in high dimensional
data, modal clustering.
Assessment of model t in high dimensional data and nonlinear space.
Statistical methodology for social sciences focusing on structural equation models.
Medical Imaging- Segmentation and characterization of anatomical objects in high dimensions
and non-linear manifolds.
Bioinformatics- focusing on classi cation of epitopes and model based clustering of microarray
gene-expression data, with applications to epitope-based vaccine development.
Teaching Experience
Spr 2008 MA 576 (Generalized Linear Models) Dept. of Mathematics and Statistics, BU.
MA 584 (Multivariate Statistical Analysis) Dept. of Mathematics and Statistics.
Fall 2007 MA 881 (Topics in High Dimensional Data Analysis) Dept. of Mathematics and
Statistics, BU.
Spr 2007 MA 576 (Generalized Linear Models) Dept. of Mathematics and Statistics, BU.
Fall 2006 MA 586 (Design of Experiments) Dept. of Mathematics and Statistics, BU.
Spr 2006 BIOS 110 ( Principles of Statistical Inference) Dept. of Biostatistics, UNC.
Fall 2005 BIOS 110 ( Principles of Statistical Inference) Dept. of Biostatistics, UNC.
Sum 2005 Taught 3 Modules in SAMSI/CRSC Undergraduate Workshop, CRSC, North Car-
olina State University.
Spr 2004 BIOS 145 (Principles of Experimental Analysis), Dept. of Biostatistics, UNC.
Fall 2001 STAT 401 (Experimental Methods), Pennsylvania State University.
Sep, 2008 2 of 7
Publications
Hong Huang Lin, Ray, S., Songsak Tongchusak, Ellis L. Reinherz, Vladimir Brusic (2008)
Evaluation of MHC class I peptide binding prediction servers: applications for vaccine
research. BMC Immunology, 9:8
Lindsay, B.G., Markatou M., Ray, S., Yang, K., Chen, S.C. (2008) Quadratic distances on
probabilities: the foundations. The Annals of Statistics Vol. 36, No. 2, page 983 1006
Ray, S., Lindsay, B.(2008). Model selection in High-Dimensions: A Quadratic-risk Based
Approach. Journal of the Royal Statistical Society - Series B Volume 70 Issue 1 (Feb), 95 118.
Ray, S., Tom Kepler (2007). Amino acid biophysical properties in the statistical prediction
of peptide-MHC class I binding. Immunome Research Oct 29;3(1):9
Li, J., Ray, S., Bruce G Lindsay. (2007) A Nonparametric Statistical Approach to Clus-
tering via Mode Identi cation Journal of Machine Learning Research 8(Aug):1687 1723.
Levy, J.H, Broadhurst R.R., Ray, S., Chaney, E.L., Pizer, S.M.(2007) Signaling local non-
credibility in an automatic segmentation pipeline Proceedings of the International Society for
Optical Engineering meetings on Medical Imaging, Volume 6512
Jeong, J., Pizer, S.M., Ray, S. (2006) Statistics on Anatomic Objects Re ecting Inter-
Object Relations. Proceedings of International Workshop on Mathematical Foundations of
Computational Anatomy.
Ray, S., Lindsay, B.(2005). The Topography of Multivariate Normal Mixtures. The Annals
of Statistics 33, 5, 2042 2065.
M. Gupta, Ray, S. (2005). Sequence pattern discovery with applications to understanding
gene regulation and vaccine design. Handbook of Statistics Ed. Chakraborty, R. and Rao,
C.R. Elsevier Press [in press]
Ray, S., Lindsay, B.,(2005) Selecting the Number of Components in a Finite Mixture: A
Risk-Based Approach. Proceedings of the of the 37th Symposium on the Interface, Computing
Science and Statistics 37.
Basu, A., Ray, S., Park, C., Basu, S. (2002) Improved Power in Multinomial Goodness-of- t
Tests, Journal of the Royal Statistical Society Series D, 51, 3, 381 393.
Ray, S. (2003) Distance-based Model-Selection with application to the Analysis of Gene
Expression Data. Electronic Thesis.
http://etda.libraries.psu.edu/theses/approved/WorldWideIndex/ETD-375/
Submitted Manuscripts
Jeong, J,, Ray, S., Han, Q., Lu X, Muller K E. and Pizer, S. M., (2008) Goodness of Predic-
tion for Principal Components of Shape: A Correlation Measure Submitted to International
Journal of Computer Vision
Sep, 2008 3 of 7
Working Papers
Ray, S. Model-based bi-clustering using two-way mixtures.
Ray, S., Marron, J.S. Feature selection based on high dimensional low sample size geometry.
Ray, S., Lindsay, B.G., Li, J. Modal EM for Mixtures and its Application in Clustering.
Ray, S., Yeong, J., Pizer, S., Muller, K., Han Q. Sample size advantages of statistics on a
nonlinear manifold to characterize nonlinear variation in a population.
Lindsay, B.G., Markatou M., Ray, S.. Degrees of Freedom in Quadratic Goodness of Fit.
Berger, J.O., Ray, S., Visser, I, Bayarri, M.J., Jang, W. Generalization of BIC.
Bollen, K.A., Ray, S., Zavisca, J. A Scaled unit information prior approximation to the
Bayes Factor.
Published software
The following softwares will be shortly available through CRAN (http://cran.r-project.org).
For current information about the packages and downloads visit http://math.bu.edu/people/sray/software/
QUADRISK: C++ binary for calculating quadratic risk of a mixture t and providing graphical
aid to high-dimensional model selection problems.
MODALITY: R-package for nding the number of modes of a multivariate normal and pro-
viding graphical and analytical representation of high-dimensional manifolds.
MHCPROP: R-package for MHC binder prediction based on biophysiochemical properties
of amino acids.
Skills
Statistics: Mixture models, Model selection in high dimensions, Asymptotics of high dimen-
sional low sample size, Bioinformatics, Immunoinformatics, Medical Image Analysis.
Programming Languages: C/C++, Perl, Python, Java, PASCAL, CSS-HTML.
Computing Platforms: Unix (Linux, Sun Solaris), DOS/Windows, Mac OS-X.
Statistical Software: Extensive experience with R/SPlus; SAS, Matlab, Mathematica, SPSS.
Recent Invited Presentations
Data Mining and Knowledge Discovery of Land Cover and Terrestrial Ecosystem Processes
from Global Remote Sensing Data NASA conference on Intelligent Data Understanding: Pre-
sented by Mark Friedl September 9-10, 2008
Modal Inference and Its Application to High-Dimensional Clustering Session on Mixture
Models: A Tool for Multilayered Clustering and Dimension Reduction at the Joint Statistical
Meetings August 3-7, 2008.
A tool for multi-layered clustering and dimension reduction. International Conference on
Sep, 2008 4 of 7
Statistical Paradigms - Recent Advances AND Reconciliations (ICSPRAR-2008), Indian Sta-
tistical Institute, Kolkata January 1-4, 2008.
Modal Inference and Its Application to High-Dimensional Clustering. Department of Bio-
statistics, University of Minnesota, October 31, 2007.
An Extended BIC for Model Selection. Joint Statistical Meetings, Salt Lake City August
1, 2007.
Modal Inference: Building the bridge between nonparametric clustering and mixture analyses.
WNAR and IMS Meetings, Irvine June 26, 2007.
Modal inference and its application to high-dimensional clustering, Department of Statistics,
Harvard University, Cambridge April 30, 2007.
Modal Inference: Building the bridge between nonparametric clustering and mixture analyses.
Department of Electrical and Computer Engineering, Boston University March 21, 2007.
Modal inference and its application to high-dimensional clustering, Department of Biotatis-
tics, Harvard University, Boston Feb 21, 2007.
Hierarchical Modal Clustering based on the Topography of Multivariate Mixtures, Interna-
tional Conference on Multivariate Statistical Methods, Kolkata, India Dec 29, 2006.
Modal inference and its application to high-dimensional clustering, Department Statistics,
University of Connecticut, Storrs Nov 9, 2006.
Modal EM for Mixtures and its Application in Clustering, Department of Mathematics and
Statistics, Boston University, Boston Sep 28, 2006.
Hierarchical Modal Clustering based on the Topography of Multivariate Mixtures, Department
of Mathematics and Statistics, Boston University, Boston Mar 22, 2006.
Hierarchical Modal Clustering based on the Topography of Multivariate Mixtures, Department
of Mathematics and Statistics, McGill University, Montreal, Canada Feb 28, 2006.
Hierarchical Modal Clustering based on the Topography of Multivariate Mixtures, Department
of Biostatistics, University of North Carolina, Chapel Hill Feb 24, 2006.
Hierarchical Modal Clustering based on the Topography of Multivariate Mixtures, Department
of Statistics, Yale University, New haven Feb 13, 2006.
Model Selection in High-Dimensions: A Quadratic risk-based approach, Department of Prob-
ability and Statistics, National University of Singapore, Singapore Feb 3, 2006
The topography of multivariate mixtures and Modal clusters, Department of Mathematics
and Statistics, University of Bristol, UK, Jan 11, 2006
Quadratic Distance:The basis for building High-dimensional model selection tool, Department
of Statistics, University of Glasgow, UK Dec 12, 2005
E ective sample size and the Bayes factor, Transition Workshop: Latent Variable Models in
the Social Sciences, SAMSI Nov 11, 2005
Model Selection in High-Dimensions: A Quadratic risk-based approach, Department of Statis-
tics, University of California, Davis Oct 6, 2005
Sep, 2008 5 of 7
Using Quadratic Risk to select models in High dimensions, Department of Statistics, London
School of Economics, London, UK Sep 27, 2005
Classi cation of MHC-I binding epitopes, WNAR/IMS Annual Meeting, Fairbanks, Alaska
June 21-24, 2005
On using Quadratic Risk to Select High dimensional Mixture Model, Annual Meeting of the
Statistical Society of Canada June 12-15, 2005
Selecting the Number of Components in a Finite Mixture: A Risk-based Approach, Joint
Annual Meeting of the Interface and the Classi cation Society of North America: Theme:
Clustering and Classi cation Washington University School of Medicine, St. Louis, Missouri
June 8-12, 2005
Bayes Factors in Structural Equation Models: Schwarz s BIC and Other Approximations,
American Sociological Association Section on Methodology: 2005 Annual Meeting, Chapel
Hill Apr 22, 2005
Selecting the Number of Components in a Finite Mixture: A Risk-based Approach, Interna-
tional Conference on the future of statistical theory, practice and education, Hyderabad, India
Dec 29-Jan 1, 2004,
The Topography of Multivariate Normal Mixtures, Seventh North American New Research
Conference Toronto, Canada Aug 4-7, 2004
Distance-based Model-selection in Mixture Distributions, International Conference on:Statistics
in Health Sciences, Nantes, France June 23-25, 2004.
Professional Activities
Organizer and chairperson of sessions in scienti c meetings.
Joint Statistical Meetings, Denver, 2008
WNAR/IMS Annual Meeting, Irvine,2007.
International Conference on Multivariate Statistical Methods, Kolkata, India, 2006
WNAR/IMS Annual Meeting, Fairbanks,2005.
Joint Annual Meeting of the Interface and the Classi cation Society of North America, 2005,
WNAR/IMS Annual Meeting, Irvine, 2007.
Organizer of NSF sponsored Undergraduate workshop in statistics, North Carolina State
University, 2005
Reviewer of several peer reviewed journal articles.
The Annals of Statistics
Journal of American Statistical Association
Journal of Royal Statistical Society (Series B)
Multivariate Analysis
Statistical Methodology
Australian and New Zealand Journal of Statistics.
Designed and taught distance learning courses at University of North Carolina, Chapel Hill.
Sep, 2008 6 of 7
Professional Memberships
2001-present American Statistical Association (ASA)
2001-present Institute of Mathematical Statistics (IMS)
2004-present Mathematical Association of America (MAA)
Student Mentoring
Burton Shank (Ph. D., Biology, Boston University) Thesis: Spatial Variation in Coral Reef
community
Role: Committee member.
Ja-Yeon Jeong (Ph.D., Computer Science, University of North Carolina at Chapel Hill) The-
sis:Estimation of probability distribution on multiple anatomical object complex
Role: Committee member.
Joshua Stough (Ph.D., Computer Science, University of North Carolina at Chapel Hill)
Thesis:Object-Relative Tissue Mixture Models for Deformable Model Segmentation
Role: Committee member.
Research Funding
NASA Carbon Cycle & Ecosystem Grant: MODIS Algorithm Re nement and Earth Science
Data Record Development for Global Land Cover and Land Cover Dynamics. NNX08AE61A
(P.I. Mark Friedl) 06/01/2008- Current Role: Statistician.
NIH Program Project Grant: Medical Image Presentation, (P.I: Pizer, S.M.) 09/15/2005-
06/31/2007 Role: Consultant.
Sep, 2008 7 of 7
Contact this candidate