Harnessing Decentralized Data to Improve Advising and Student Success - NASPA...
CV_Shibasish
1. Dr. Shibasish Dasgupta
Contact
Information
3126 FSB Cell: (001) 352-871-7282 , Office: (001) 513-529-3212
800 E. High Street Email: shibasish.dasgupta@gmail.com
Oxford, OH 45056-3600, USA www.linkedin.com/pub/shibasish-dasgupta/6/28/81a
Research
Interests
Statistical Methods for Selection and Evaluation of Biomarker, Statistical Learning, in particular,
Variable Selection Procedures like LASSO and its direct descendants, Covariance Estimation,
Predictive Modeling, and Bayesian Theory, more specifically, Asymptotic Expansion of the Posterior
Density and Posterior Consistency in High Dimensional regimes.
Current
Position
Visiting Assistant Professor, Department of Information Systems and Analytics
Farmer School of Business, Miami University
Oxford, Ohio, USA August 2014 – present
Education Fred Hutchinson Cancer Research Center, Seattle, Washington, USA
Post-Doctoral Research Fellow at the Population Sciences
Statistics Department September 2013 – August 2014
• Research Focus: Statistical Methods for Selection and Evaluation of Biomarker
University of Florida, Gainesville, Florida, USA
Doctor of Philosophy in Statistics August 2007 – August 2013
• Dissertation Title: High Dimensional Inference and Variable Selection
• Adviser: Prof. Malay Ghosh, Co-adviser: Prof. Kshitij Khare
University of Pune, Pune, India
(In collaboration with TATA Research Development and Design Center, Pune, India)
Master of Philosophy in Statistics August 2004 – October 2005
• Dissertation Title: Application of Support Vector Machines in Classification Problems
• Adviser : Prof. M.B. Rajarshi, Co-adviser: Dr. Aniruddha Pant
University of Pune, Pune, India
Master of Science in Statistics August 2002 – June 2004
St. Xavier’s College, University of Calcutta, Calcutta, India
Bachelor of Science in Statistics August 1999 – June 2002
Honors and
Awards
Travel Award for the International Society for Bayesian Analysis (ISBA) World
Meeting, Cancun, Mexico, ISBA, July 2014.
Student-Postdoc Advisory Committee (SPAC) Course Scholarship Award, Fred
Hutchinson Cancer Research Center, February 2014.
Travel Award for the Eastern North American Region (ENAR) Meeting, Orlando,
Department of Statistics, University of Florida, March 2013.
College of Liberal Arts and Sciences (CLAS) Dean’s Travel Award for the Joint Statis-
tical Meetings (JSM), San Diego, University of Florida, August 2012.
Chaired a session on Bayesian Methods for Clinical Trials, JSM, San Diego, August 2012.
Travel Award for the Conference Board of the Mathematical Sciences (CBMS) Confer-
ence on Model Uncertainty and Multiplicity, University of California, Santa Cruz, July 2012.
Graduate Student Travel Award for JSM, Miami, University of Florida, August 2011.
Chaired a session on Bayesian Computation, JSM, Miami, August 2011.
2. One of the top students in the PhD Qualifying Examination, Department of Statistics,
University of Florida, 2009.
Travel Award for the Conference on Frontiers of Probability and Statistical Science,
University of Connecticut, Storrs, May 2008.
Winner of Indian Society for Probability and Statistics (ISPS) project competition in
MS, India, February 2004.
One of the top students in MS, University of Pune, India, Aug 2002 - June 2004.
University Grant Commission (UGC) and Pfizer scholarships in MS, India, 2002 - 2004.
National Scholarships, Secondary and Higher Secondary Examinations, India.
Publications Dasgupta, S. (2015). High Dimensional Posterior Consistency of the Bayesian Lasso. To appear in
Communications in Statistics – Theory and Methods.
Dasgupta, S. (2015). Variable Selection Using Kullback-Leibler Divergence Loss. To appear in
Journal of the Indian Statistical Association.
Dasgupta, S., Khare, K. and Ghosh, M. (2014). Asymptotic Expansion of the Posterior Density in
High Dimensional Generalized Linear Models. Journal of Multivariate Analysis, 131, 126–148.
Dasgupta, S., Sinha, S. and Huang, Y. (2015). A Bayesian Predictive Approach for Designing
Biomarker Validation Studies. (in progress)
Huang, Y., Dasgupta, S. and Wang, P. (2015). Joint Variable Selection for Risk Predictors and
Baseline Predictors in Two-Phase Sampling. (in progress)
Dasgupta, S. (2015). The Bayesian Nested Lasso. (in progress)
Teaching
Experience
Department of Information Systems and Analytics, Miami University, Oxford, Ohio, USA
Visiting Assistant Professor Fall 2014 – Spring 2015
Teaching Business Statistics.
Department of Statistics, University of Florida, Gainesville, Florida, USA
Teaching Instructor Spring 2010 – Fall 2011
Teaching instructor for the courses Introduction to the Practice of Statistics II (huge classes of more
than 200 students each!), Introduction to Probability/Fundamentals of Probability (this course is
intended for undergraduate as well as graduate students from different disciplines) and Engineering
Statistics.
Department of Statistics, University of Florida, Gainesville, Florida, USA
Teaching Assistant August 2007 – June 2013
Teaching assistant for undergraduate and graduate courses in Statistics, including Introduction to
the Practice of Statistics I and II, Engineering Statistics, Non-parametric Statistical Methods,
Fundamentals of Probability and Statistical Methods for Research II.
Department of Statistics, University of Pune, Pune, India
Teaching Assistant August 2004 – December 2005
Teaching assistant for graduate courses in Statistics, including Mathematical Analysis and Data
Mining while pursuing “Master of Philosophy” research.
Industrial
Experience
Persistent Systems Ltd. (PSL), Pune, India
(A Leading Provider of Outsourced Software Product Development Services and One of the Best
Leaders in Research and Development Services)
Module Lead June 2006 – May 2007
• Worked in the Statistical Learning and Data Mining group called “Analytics Centre of Excel-
lence.” Involved in various projects, one of which is named “Snippet Tagging Using Text Mining.”
3. This was a unique text classification problem posed by Gridstone Research which has a powerful and
flexible analysis and modeling platform that helps investment professionals make better investment
decisions.
• Used hierarchical multi-class classification using different statistical machine learning techniques
like SVMs, Naive Bayes and Hidden Markov Models in various predictive analytics projects and one
of our biggest clients was SPSS.
TATA Research Development and Design Center (TRDDC), Pune, India
A division of Tata Consultancy Services (TCS)
(Asia’s Largest Consulting Company)
Scientist October 2004 – May 2006
• Worked in the “Business Analytics R&D” group. Pursued M.Phil. research in collaboration
with TRDDC in the mathematical and statistical aspects of Support Vector Machines (SVMs), its
related areas and its application in the classification problems. Also explored the possible applications
of this newly and fast emerging area in finance, medical science and business.
• Involved in a Customer Relationship Management (CRM) project to predict customer churn by
different modeling tools like Logistic Regression, SVMs, Decision Trees and survival data mining.
My other works at TRDDC are the following:
• Variable Selection Methods: Principal Component Analysis, Correlation and Information Theory.
• Inventum: A Predictive Analytic Modeling of Workers Compensation Classification (a project about
two-class classification problem).
• Choosing Multiple Parameters for SVM: Automatic parameter tuning of SVM (based on the
theory of minimizing the upper bound of the generalization error of a classification problem).
• Automated Rail Weight Identification System: A project on multi-class classification problem.
• Feature selection for SVM using Automatic Tuning.
• Explored a new area for very large SVM training using Core Vector Machines (CVMs), its theory
and applications.
Talks and
Presentations
A Bayesian Approach to Design Future Studies for Comparing Biomarkers. Conference on Recent
Advances In Statistics, Department of Statistics & Center for Advanced Studies, University of Pune,
INDIA, January 2-3, 2015.
High Dimensional Posterior Consistency of the Bayesian Lasso. ISBA World Meeting, Cancun,
Mexico, July 2014.
A Bayesian Approach to Design Future Studies for Comparing Biomarkers. Frontiers of Hierarchical
Modeling in Observational Studies, Complex Surveys and Big Data: A Conference Honoring Professor
Malay Ghosh, University of Maryland, College Park, USA, May 2014.
High Dimensional Inference and Variable Selection. PhD Defense, Department of Statistics,
University of Florida, Gainesville, USA, June 2013.
Variable Selection with Kullback-Leibler Divergence Loss. Novartis Oncology, NJ, USA, June 2013.
Variable Selection: A Journey. JSM, San Diego, USA, August, 2012.
Probabilistic Modeling of Text Data: A Review. JSM, Miami, USA, July, 2011.
An Introduction to the Probabilistic Modeling of Text. Department of Statistics, University of Florida,
Gainesville, USA, September 2010.
Support Vector Machines: A Useful Tool for Classification. IISA Conference on Frontiers of
Probability and Statistical Science, Organized by the Department of Statistics, University of
Connecticut, Storrs, USA, May 2008.
Two Approaches to Look at Support Vector Classification and its Application. Joint Statistical
Meeting of the IISA and International Conference on Statistics, Probability and Related Areas,
Organized by the Department of Statistics, Cochin University of Science and Technology, Cochin,
India, January 2007.
4. Application of Support Vector Machines in Hierarchical Classification in a Text Mining Problem.
Department of Statistics and Center for Advanced Studies, University of Pune, India, September
2006.
A Comparison of Model Selection Methods for Support Vector Machines. National Conference on
Statistical Inference organized by the Department of Statistics and Center for Advanced Studies,
University of Pune, India, January, 2006. (An updated version of this paper including results of
Genetic Algorithm was also accepted in the 2006 International Conference on Data Mining DMIN06,
June 2006, Las Vegas, USA.)
Application of Support Vector Machines in Classification Problems. TATA Research
Development and Design Center, Pune, India, December 2005.
A Guide to Support Vector Machines. Department of Statistics and Center for Advanced Studies,
University of Pune, India, March 2005.
Comparison of Fisher’s and Support Vector Machines for Classification. TATA Research
Development and Design Center, Pune, India, December 2004.
Courses
Taken
Introduction to Theoretical Statistics I and II , Regression Analysis, Basic Design and Analysis of
Experiments, Theory of Linear Models, Generalized Linear Models, Elements of Statistical Learning
Theory, Statistical Inference, Variable Selection, Survival Analysis, Bayesian Theory, Nonparametric
Statistics, Probability Theory I and II, Limit Theory, Advanced Inference, Machine Learning.
Technical
Skills
R, SAS, MATLAB, SPSS, MINITAB, LATEX, MS Office.
Professional
Memberships
• American Statistical Association (ASA)
• Institute of Mathematical Statistics (IMS)
• International Society for Bayesian Analysis (ISBA)
• International Indian Statistical Association (IISA)
References Available on request.