James has provided statistical expertise to a variety of projects, particularly in biomedical domains. While a post-doc at the UM Cancer Center, he extensively used advanced bioinformatic and statistical tools for large-scale data analyses. His research interests include structure learning for biological networks, especially from perturbation or time-series data, and enrichment analysis. Other areas of expertise include: regression modeling, graphical models, high performance computing, inter-rater reliability, and parameter estimation for dynamic systems.
Josh has academic degrees in both Statistics and Computer Science, with deep experience in statistical methodology and use of statistical software. His areas of interest include causal inference, propensity score matching, multiple regression diagnosis, structural equation modeling, software implementation and general computing concerns.
I am interested in geospatial science and its application for spatiotemporal analyses, policy, and decision-making. I have nearly fifteen years of experience in combining observations from multiple satellite and airborne sensors, vector data from secondary and primary sources, and models to monitor and understand spatiotemporal variation across a range of topics. My specific interests include geospatial analysis, image processing, location analysis, spatially distributed dynamic models, model inversion, radiative transfer modeling, signal processing, and nonlinear dynamics. In my research and consulting I have analyzed a variety of data including remotely sensed data such as from MODIS, Landsat, AVHRR, OCO-2, AIRS, AVIRIS, and SRTM; vector data such as from Census, road network, FLUXNET, SURFRED, OzFlux, Phenocam, and weather stations; and gridded and reanalysis data such as from CMIP5, NCEP, NOAA, and NASA. I often combine multiple languages including MATLAB, R, and Python, extensively use ArcGIS and QGIS, and occasionally also turn to Mathematica and ENVI/IDL. I have assimilated data across multiple spatial and temporal scales to bridge gaps in our understanding of small-scale processes and large-scale patterns. My domain knowledge includes remote sensing, mass and energy exchange between the land surface and the atmosphere, photosynthesis, carbon and water cycle, ecosystem ecology, and watershed analyses. I have set up and used instruments such as radiometer, spectrometer, gas exchange, and PAR sensors. A list of my publications is available at https://www.researchgate.net/profile/Manish_Verma21.
Before coming to research/academia I worked for more than a decade with non-governmental organizations promoting better and inclusive management of natural resources in ecologically fragile areas. I also worked as a consultant and advised several organizations including Oxfam, KfW, Winrock, and Aga Khan Rural Support Program. I love wild animals and go through phases where my enchantment skips from birds to reptiles to mammals. I like reading, cooking, seeking connections between abstract ideas and realities, and traveling with my spouse (Preeti) and my son (Aman).
Kerby has collaborated with researchers in many areas of natural and social science, with a particular focus on studies involving high dimensional biological data, including genomics, biological imaging, statistical genetics, chemical informatics, health outcomes, and medical claims data. He also has extensive experience in software development, primarily in Python and Go. His research focuses on computational statistics, and methods for modeling complex dependent data including multilevel, longitudinal, and spatial/temporal data.
Shyamala has many years of experience with study design and statistical analysis, with a focus on time series analysis, design of experiments, and regression modeling. She has worked in a variety of disciplines, especially in the social sciences.
Michael has modeling and consulting experience across varied domains, including social and physical sciences, business, and humanities. He has particular expertise in mixed models, additive models, latent variable models, text analysis, structural equations modeling, machine learning techniques, Bayesian inference, reproducible research practices, data visualization, and efficient programming practices.
For more about Michael, visit https://m-clark.github.io/.
Chris is a Statistician in the U-M Department of Ophthalmology and Visual Sciences and formerly was faculty at Oberlin College and the State University of New York at Buffalo. He currently is the lead statistician for the Northeast Regional Core of the Women’s Health Initiative and has additional experience as a statistical consultant for Roswell Park Cancer Institute. His areas of expertise include survival analysis, linear and generalized linear modeling, missing data methods, metabolomics, and analysis of administrative claims data.