Shane T. Jensen
Department of Statistics
The Wharton School
University of Pennsylvania
463 Huntsman Hall
3730 Walnut Street
stjensen at wharton.upenn.edu
My research program is a collaborative and multidisciplinary effort that spans the fields of architecture, urban planning, criminology and statistics. Our endeavor is to take the available data on cities and set up artificial experimental situations that allow us to learn as objectively as possible about what aspects of city environments are associated with safety and other outcomes.
You can read more about our data collection and analysis pipeline in our paper:
Analysis of Urban Vibrancy and Safety in Philadelphia by C. Humphrey, S.T. Jensen, D. Small and R. Thurston. arXiv: 1702.07909.
You can also read more about our goals in these media articles:
Knowledge@Wharton: How Urban Planners Can Encourage Vibrancy — and Create Safer Cities. This article also has a podcast link to an interview I did on the Wharton Business radio show "Dollars and Change".
Our urban analytics research program has received generous support from the Wharton Social Impact Initiative.
|Other Research Interests|
1. Genetics and Molecular Biology
Developing sophisticated statistical models for the evolution of genomic sequences. Areas of application include the response of HIV under various therapies as well as evolution during cancer progression. Developing models for combining heterogeneous data sources to refine predictions about co-regulated genes and regulatory networks in cells.
2. Statistics in Sports
Developing novel statistical models for the comparison of baseball players in terms of on-field performance. Evaluation of fielding ability as well as prediction of future hitting and pitching performance. Quantifying player performance in hockey.
3. Bayesian Nonparametrics
Extensions of Dirichlet processes for grouped and ordered data. Alternative prior processes for non-parametric clustering. Tree-based approaches for high-dimensional settings.
4. Economics and Marketing
Estimating income volatility while allowing for heterogeneity over time and between individuals in the population. Exploring the relationship between income volatility and risk aversion. Modeling career choice as a function of risk aversion. Models for missing data in marketing research.
|Individual Project Pages|
SAFE: Spatial Aggregate Fielding Evaluation, our methodology for measuring fielding ability in major league baseball players using a hierarchical probit model. Results are presented across seven seasons of high-resolution ball-in-play data.
COGRIM: Bayesian variable selection model for regulatory network inference through the integration of gene expression data, ChIP binding data and sequence motif data.
PHYLOCLUS: Suite of perl programs for clustering co-regulated genes based on phylogenetically discovered transcription factor binding motifs.
MOTIF CLUSTERING: Perl programs and supplemental material for clustering transcription factor binding motif matrices based on a hierarchical Bayesian model.
|Past Media Attention|
2015/10/13: Articles about my collaboration with researchers investigating why elephants get less cancer than humans: Newsweek and IFLscience
2008/02/17: Media Attention for our Baseball Fielding Research: AP Boston Globe Popular Science Wired Science Citizen
2008/02/11: Media/Blog Discussion of our NY Times article: ESPN article Freakonomics Blog Statistics Blog