You are in the College of Science and Liberal ArtsCollege of Science and Liberal Arts

Department of Mathematical Sciences

Dhar, Sunil K.

Contact Info
Title: Professor
Email: sunil.k.dhar@njit.edu
Office: 211B Cullimore
Hours: Spring 2014: Mondays 1:55 pm to 3:55 pm and Tuesdays 4 pm to 5 pm.
Phone: 973-596-3488
Dept: Department of Mathematical Sciences
Webpage: http://web.njit.edu/~dhar/courses.html

Academic Interests: model building, inference

About Me

The research focus of Sunil Dhar has been on model building and inference. The inverse sampling scheme’s multivariate distribution was generalized to the new Log-linear Modeling, Soumi and Dhar (2008) and (Dhar; 1995) under the generalized inverse sampling scheme. This new multivariate distribution helps generalize other “order-k” distributions (Dhar 1996). His ongoing research involves proving existence, computing and developing robust and efficient minimum distance estimators such as L2-distance type, under the following models: linear, AR [k], the additive effects outliers, and the two-sample location model, Dhar and Chen (2005). He also developed functional least squares estimators under the additive effects outliers’ model. An optimization technique for the general class of sums of absolute multivariate linear functional has been developed by him. He has acquired statistical consulting experience and applies it to the health sciences. Constructing new bootstrap-based test for establishing noninferiority in clinical trials and he has developed methods training clinical staff in pharmaceutical industry to do biostatistical research, Dhar and Kianifard (2006).

Education

  • PhD, Statistics, Michigan State University
  • MS, Statistics & Probability, Michigan State University
  • MSc, Mathematics (First Class), University of Poona
  • BSc, Mathematics (First Class), University of Poona

Courses I Teach

Not teaching any courses this semester

Courses Taught


Research Interests

  • Modeling 
    • Multivariate Count
    • Pharmaceutical Statistical
    • Reliability
    • Survival
  • Inference
    • Non parametric
    • Robust statistics
The research focus of Sunil Dhar has been on model building and inference. The inverse sampling scheme’s multivariate distribution was generalized to the new Log-linear Modeling, Soumi and Dhar (2008) and (Dhar; 1995) under the generalized inverse sampling scheme. This new multivariate distribution helps generalize other “order-k” distributions (Dhar 1996). His ongoing research involves proving existence, computing and developing robust and efficient minimum distance estimators such as L2-distance type, under the following models: linear, AR [k], the additive effects outliers, and the two-sample location model, Dhar and Chen (2005). He also developed functional least squares estimators under the additive effects outliers’ model.  An optimization technique for the general class of sums of absolute multivariate linear functional has been developed by him.  He has acquired statistical consulting experience and applies it to the health sciences. Constructing new bootstrap-based test for establishing noninferiority in clinical trials and he has developed methods training clinical staff in pharmaceutical industry to do biostatistical research, Dhar and Kianifard (2006).

 

Currest Research

Ongoing research in count data modeling, lifetime reliability models involve deriving new multivariate models with demonstrated applications. Pharmaceutical statistical analysis under clinical trials is among his current research.


Selected Publications

  • Log-linear Modeling under Generalized Inverse Sampling Scheme: Communications in Statistics - Theory and Methods, Volume 37(8), 2008, pp. 1237-1244 (with Ph.D. student Soumi Lahiri)
  • "Computation of Certain Minimum Distance Estimators in AR[K] Model," Journal of American Statistical Association, 88(1/21), 1993, pp 278-283.
  • "Minimum Distance Estimation in an Additive Effects Outliers Model," The Annals of Statistics, 19(1), 1991, pp 205-228.  
  •   More Publications