Characterizing Morphologies of Premalignant Fibroblast Cells [Paper]
Theresa Gebert, Larry Wasserman

Main Idea: While cancerous cells have been studied extensively, there is little understanding of the subtle differences between healthy cells and cells immediately prior to cancer. We analyze single-cell-level data derived from patients with and without a genetic predisposition to cancer. We use LASSO, SVMs, and random forests to identify what features of cell shape consistently appear in cells with the pre-cancerous mutation and benchmark our performance against convolutional neural networks. Future work will use these to discover biological pathways relevant to cancer and candidates for drug targeting.

Textbook: Introduction to Stochastic Processes with R [Manuscript] [Code]
Theresa Gebert, Natesh Pillai

Main Idea: This 300-page book is intended as a first course in stochastic processes for advanced undergraduate students or graduate students. Despite existing texts on stochastic processes, few of these texts incorporate code into their material in a meaningful way. We believe simulations and algorithms (e.g. Gibbs sampling) are the building blocks of research in statistics. Rather than approaching this material from a purely mathematically derivative perspective, we reinforce understanding through verifications via simulation as well. We also incorporate material rarely seen at an undergraduate level, such as martingales and stopping time theory. Every chapter concludes with theoretical and simulation-based exercises.

Identifying Perspective-Taking Failures in Physical vs. Mental States [Paper]
Matthew Jordan, Theresa Gebert, Christine Looser

Main Idea: Accurately inferring the values and preferences of others is crucial for successful social interactions. Nevertheless, without direct access to others’ minds, perspective taking errors are common. Across four studies we demonstrate a systematic perspective-taking failure: people value their minds more than their bodies, but fail to realize others share those values, often believing that others value their bodies more than their minds. This bias has implications for the ways in which we create social policy, judge others’ actions, make choices on behalf of others, and allocate resources to the physically and mentally ill.


Advanced Methods for Data Analysis (Spring 2018)
Ann Lee, Carnegie Mellon University

Statistical Computing (Fall 2017)
Peter Freeman, Carnegie Mellon University

Introduction to Stochastic Processes (Spring 2015)
Natesh Pillai, Harvard University

Introduction to Probability (Fall 2014)
Joe Blitzstein, Harvard University

Introduction to Theoretical Statistics (Spring 2014)
Kevin Rader, Harvard University

Introduction to Probability (Fall 2013)
Joe Blitzstein, Harvard University


I was born in Germany, grew up in Princeton and Bavaria, and currently split my time between Pittsburgh and San Francisco. I am passionate about encouraging underrepresented demographics to pursue Statistics, and I volunteer as a statistical consultant for non-profit organizations. I also enjoy weight lifting, hiking, painting, and traveling.

GitHub, LinkedIn, Art Portfolio and E-mail (theresa['at']stat[.]cmu[.]edu).