Director and Research Professor
Survey Research Center, Institute
for Social Research
Professor of Biostatistics, School
of Public Health
University
of Michigan
Research Professor, Joint Program in
Survey Methodology, University of Maryland
I was born in Nagpur
, the central part of India. I had my schooling at Saraswati Vidyalaya. I
received my Bachelors and Masters degrees from Institute
of
Science, Nagpur
University . My Bachelor's degree was in Mathematics,
Physics and
Statistics and Masters in Statistics. I
worked as a lecturer of Statistics at the Institute
of Science for three
years.
I received a Masters degree in
Statistics from Miami
University, Oxford,
Ohio. I received Ph.D. in Statistics from Harvard
University. My
Ph. D. supervisor was Don Rubin and the committee members were Arthur
Dempster and Nan Laird.
I joined University of Washington
as an Assistant Professor in 1987. I moved to University
of Michigan as Associate
Professor in 1994 with a joint appointments in the School
of Public Health and the Institute
for Social Research. I
served as Director of Survey Methodology Program, at the Institute
for Social Research from 1997 to 2001. I became a Professor in 2002
and was Chair, Department of Biostatistics 2010-2014. I became the
Director of Survey Research Center, Institute for Social Research in
2014.I am also a Research
Professor at the Joint Program in Survey Methodology, University of
Maryland.
My research interests span several areas in statistics including
Bayesian Methods, Survey Design and Analysis, Missing Data, Disclosure
Limitation, Measurement Error, Spatial Statistics, Longitudinal Data,
Small Area Estimation, Combining Information from Multiple Sources and
Statistical Methods for Epidemiology. For more information about my publications visit
Google
Scholar Page.
I have developed a software, IVEWARE, for multiple
imputation analysis which is available in both Linux and Windows
platforms as a SAS add-on or stand-alone. Download, Use
and Send Comments. It is free.
I have published a book, Missing Data Analysis in Practice, which provides
practical methods for analyzing missing data along with the heuristic reasoning for understanding the theoretical underpinnings.
At University of Washington, I collaborated with Researchers at the Cardiovascular Health Research Unit and Cancer Prevention Research Program, Fred Hutchinson Cancer Research Center. At Michigan, I direct Biostatistics Collaborative and Methodology Research Core (BCMRC), a research unit designed to foster collaborative and methodological research with the researchers in other departments in the School of Public Health and other allied schools. I am the Director of Biostatistics and Measurement Core for Michigan CTSA housed at Michigan Institue for Clinical and Health Research. I am an Associate Director of the Center for Research on Ethnicity, Culture and Health (CRECH). A faculty member at the Center of Social Epidemiology and Population Health (CSEPH) and also affiliated with the University of Michigan Transportation Research Institute (UMTRI).
I follow a Bayesian approach for
statistical
inference based on statistical
models but use frequentist ideas for model checking and
model diagnostics. I view that randomization (or near randomization in
observational studies) is essential for an objective data collection.
For most practical purposes, the likelihood function based on a
well-researched and well-formulated model should dominate the prior
distribution when drawing inferences. Therefore, the prior
distributions are
usually diffuse (justifying the expense of data collection!) but I will
not hesitate to use non-diffuse priors, if based on solid empirical
evidence.
cv, unpublished
papers, courses
taught and
other interests.