Priya Kohli


Priya Kohli

Associate Professor of Statistics
Assistant Chair of the Mathematics and Statistics Department

Joined Connecticut College: 2012

Education
M.S., Indian Agricultural Statistical Research Institute, New Delhi
M.S., Northern Illinois University
Ph.D., Texas A&M University


Specializations

Covariance Modeling

Longitudinal Studies

Multivariate Studies

Time Series Analysis

Priya Kohli specializes in the areas of covariance modeling, longitudinal studies, multivariate studies, missing data, and time series modeling. She also works in interdisciplinary research areas including RNA-seq analysis, healthcare, and environmental sciences.

Professor Kohli’s research accomplishments include publications in some of the most distinguished international statistics and interdisciplinary journals. Her research work has been accepted as a U.S. patent and a European patent. She has been invited as a speaker to present her research work at several prestigious statistics and data science conferences.

More recently, she has been working on developing visualization and modeling tools for mean and covariance estimation in multivariate longitudinal analysis. Multivariate longitudinal studies are common in a diverse array of fields, including biostatistics, clinical trials, epidemiology, genetics, and public health. In such studies, measurements from multiple outcomes are recorded repeatedly over time. The covariance matrix of multiple outcomes includes important information about temporal-dependencies between the repeated measurements and cross-dependencies between outcomes at contemporaneous and non-contemporaneous times. Joint mean-covariance modeling of multivariate longitudinal data helps in understanding the trends and dependence patterns among repeatedly measured outcomes. Accurate covariance modeling improves the accuracy of the estimated mean and provides scientific information that is useful for making valid statistical inferences. Graphical and data-driven tools that can aid in visualizing the dependence patterns among multiple longitudinal outcomes are limited. Kohli introduced a new software package in R called MLGM for Multivariate Longitudinal Graphical Models. In addition to visualizing the existing patterns, MLGM depicts modeling the mean and covariance for multiple longitudinal outcomes measured at regular or irregular time points using least-squares estimation.

Kohli is currently collaborating to work on estimating the pre-conditions of a country required for wide usage and a practical impact of mobile money in the presence of a public health crisis in different countries. Part of the pandemic related economic slowdown in developing countries, would occur simply as a result of people being too afraid to venture out into the market to make purchases due to health risks from in-person transactions and handling cash money. This is the result of low financial inclusion and a large dependence on cash transactions. The objective work is to draw useful lessons from the past to provide practical policy direction for digital payment platforms in the world today.

Kohli teaches Time Series Analysis, Statistical Computing with R, Probability and Statistical Model, Statistical Consulting, and Mathematical Statistics.

Recent publications:

1. Kohli, P., Du, S.*, and Shen, H.* (2021). Graphical models for mean and covariance of multivariate longitudinal data. Statistics in Medicine, 15;40(23):4977-4995. R package MLGM.

2. Marazzi, L.*, Kohli, P., and Eastman, D. (2020). Transcriptome dataset for RNA-seq analysis of axolotl embryonic oropharyngeal endoderm explants. Data in Brief, 32, 106126.

3. Kohli, P., Marazzi, L.*, and Eastman, D. (2020). Transcriptome analysis of axolotl oropharyngeal explants during taste bud differentiation stage. Mechanisms of Development. 161, 103597.

4. Lopez-Anuarbe M. and Kohli P. (2019).Understanding male caregivers' emotional, financial, and physical burden in the United States. Healthcare, 7(2). E72.

5. Kohli P., Siver P. A., Marsicano L., Hamer J. and Coffin A*. (2017). Assessment of long-term trends for management of Candlewood lake, Connecticut, USA. Lake and Reservoir Management, 33(3), 280-300.

6. Harvill J. L., Kohli P., and Ravishanker N. (2017). Clustering nonlinear, nonstationary time series using BSLEX. Methodology and Computing in Applied Probability, 19(3), 935-955.

Here * denotes an undergraduate student.

Online:

R Package MLGM https://priyakohli5.github.io/MLGM/

View my CV

Majoring in Statistics and Data Science

Majoring in Mathematics.

Visit the department of mathematics and statistics website.

Contact Priya Kohli

Mailing Address

Priya Kohli
Connecticut College
Box # MATHEMATICS & STATISTICS/Fanning Hall
270 Mohegan Ave.
New London, CT 06320

Office

312 Fanning Hall