The Lindsay Visiting Assistant Professor program was created in 2015 to honor Professor Bruce Lindsay, a leading statistician, mentor and a long-time faculty member of Penn State Statistics.
Current Lindsay Assistant Professors
Zhou Lan joined the Department of Statistics at Penn State University as a Lindsay Visiting Assistant Professor in 2019. Before joining Penn State Statistics, he obtained a Ph.D. degree in Statistics in the Department of Statistics at North Carolina State University and a Master of Science in Statistics in the School of Mathematics at Georgia Institute of Technology. He is the winner of North Carolina State University Paige Plagge Award. This award is given for good citizenship to "a graduate student with an outstanding academic record, who in the judgment of the committee has especially enhanced the life of fellow students with encouragement, generosity and/or humor."
Zhou’s primary research interests are in the areas of Bayesian nonparametrics, Bayesian methods, spatial statistics, spatial random fields, and neuroimaging. His major contribution is "spatial modeling of positive definite matrices, with applications to diffusion tensor imaging (DTI)". DTI is a popular neuroimaging tool in revealing the brain’s tissue structure. DTI maps and characterizes the 3-D diffusion of water molecules as a function of the spatial location. The diffusion process in the brain reflects interactions with many obstacles, such as fibers, thereby revealing microscopic details about the underlying tissue architecture. Unlike ordinary images where scalars are summarized for each voxel, a distinguishing feature of DTI is each voxel is associated with a 3 × 3 symmetric positive definite matrices. Spatial statistical analysis of DTI data is challenging due to the difficulty of modeling positive matrix-variate responses. We proposed two approaches: a spatial Wishart process (a continuous random field) and a Potts model (a discrete random field). Both are spatial random fields supporting (inverse) Wishart matrices. The associated papers won the 2019 ICSA Applied Statistics Symposium Student Paper Competition and 2019 ASA Student Paper Competition.
At Penn State Statistics, he will be continuing working on spatial statistics, epidemiology, disease mapping, and statistical inferences of positive definite matrices.
Roberto obtained a Master’s degree in International Affairs at the LUISS Guido Carli University in Rome. He served subsequently as a consultant for the United Nations Economic Commission for Europe (UNECE), the Organization for Economic Cooperation and Development (OECD) and the accounting firm of Ernst & Young. Molinari subsequently received his M.Sc. and Ph.D. in Statistics at the University of Geneva, Switzerland, and then became a Visiting Assistant Professor in Statistics at the University of California, Santa Barbara.
Molinari then spent a year in Senegal where he served as a statistical consultant for the United Nations International Children's Emergency Fund (UNICEF); as a researcher at the Global Research and Advocacy Group (GRAG) on marginalized communities and the non-profit organizations (GRAG), conducting research on marginalized communities in sub-Saharan Africa; and as a researcher in the field of climatology and epidemiology at the Center for International Research on Environment and Development (CIRAD).
Molinari’s research interests include robust statistics, stochastic processes, model selection in high dimensions, computational statistics, and data privacy. In his position as a Lindsay Assistant Professor, Molinari is researching a variety of topics among which are the development of non-parametric bootstrap techniques for the creation of differentially-private synthetic data with Prof. Aleksandra Slavkovic and Michelle Pistner, a doctoral student; in this project, he is also researching new criteria to assess the level of privacy of statistical procedures and synthetic data. Molinari also is conducting research with co-authors at Penn State (Prof. Stephane Guerrier and Dr. Mucyo Karemera), and with colleagues at the University of Illinois (Urbana-Champaign), the University of Maine, UC-Santa Barbara, the University of Geneva, the École Polytechnique Fédérale de Lausanne (EPFL-Switzerland), and CIRAD (Senegal).
Molinari is now finalizing new algorithms for the robust estimation of a large class of time series models, to be implemented in the gmwm-R package available on GitHub, and he is extending its applicability to multivariate and non-stationary time series analysis. Molinari is also developing new approaches for gene selection problems, Granger-causality for biological problems, and computationally efficient approaches to estimate complex models in high dimensional settings, with the corresponding implementation in R packages.
Danning received in 2013 her Ph.D. in Statistics from the University of Minnesota, Minneapolis, advised by Professor Tiefeng Jiang. From 2013-2015, she was a post-doctoral research associate under the supervision of Professor Richard Samworth at the University of Cambridge, England.
Li worked subsequently as an Assistant Professor of Statistics at Jilin University, China and then in 2017 she joined the Department of Statistics at Penn State as a Lindsay Visiting Assistant Professor.
Li’s research areas include random matrix theory, high-dimensional statistical inference, empirical process, and applications to biological science, network science, and social science. Her papers have been published in journals in probability and statistics, including the Journal of Theoretical Probability, the Journal of Mathematical Physics, the Institute of Electrical and Electronics Engineers (IEEE) Transactions on Information Theory, and Statistica Sinica.
Li’s publications include results on the spectra of truncated random unitary matrices; these results are germane to applications involving quantum systems with absorbing boundaries, optical and semiconductor superlattices, quantum conductance, and the distribution of resonances for open quantum maps. She has also published research on the smallest eigenvalues of random matrices, a problem which arises in electrical engineering and in multivariate statistical analysis; her research connects the limiting distributions of the smallest eigenvalues with the celebrated Tracy-Widom probability distributions. Most recently, Li has obtained results that connect Stein’s method of unbiased risk estimation with the estimation of high-dimensional covariance matrices.
In current research, Li is now working with her co-authors to develop power enhancement tests for high-dimensional means, covariance matrices, regression models, and related network models.
Former Lindsay Assistant Professors
Cremona joined the Department of Statistics at Penn State as a post-doctoral researcher in 2016, working with Francesca Chiaromonte and Kateryna D. Makova, and in 2017 she was appointed to a Lindsay Visiting Assistant Professorship. Marzia Cremona received her B.Sc. in 2009, and M.Sc. in 2011, in Mathematics from the Università degli Studi di Milano, Italy. She obtained in 2016 her Ph.D. in Mathematical Models and Methods in Engineering from the Politecnico di Milano, with a thesis entitled "Statistical methods for omics Data." Her doctoral research advisor was Prof. Piercesare Secchi with co-advisors Prof. Laura M. Sangalli and Prof. Simone Vantini.
Cremona’s primary research interests are in the areas of statistical learning, computational statistics, and statistical “omics”. In her research, Cremona develops statistical and computational methods for the analysis of large, high-dimensional, and complex data – in particular, functional data. An important aspect of her research is its collaborative and multidisciplinary nature; indeed, much of her work is at the interface of statistics and computational biology, and her main application areas are the biomedical and “omics” sciences.
Among her publications are articles on the clustering of ChIP-seq data using peak shape or shape indices, the genome-wide effects of non-B DNA on polymerization speed and error rate, and the influence of the genomic landscape on the integration and fixation of endogenous retroviruses. In research in other areas, she has also studied methods for predicting railway wheel wear using kriging methods.
In research that was completed recently, Cremona has announced results on high-resolution views of adaptive events and on probabilistic methods for clustering and motif discovery in functional data.