Department of Statistics and the College
Rina Foygel Barber’s research pertains to the development and analysis of methods that attempt to uncover hidden patterns and structure in data sets that might be large, incomplete, messy, or uninterpretable. Her approach is primarily theoretical, focusing on the mathematics behind these problems, while also testing methods on real data. Her work seeks to answer questions relating to the types of new methods that can be developed in order to improve the ability to extract information from data, the expected levels of performance of existing methods, and trade-offs involving accuracy. Among the problems that are the focus of her research are sparse regression and model selection, sparse graphical models, low rank matrix estimation, and the trade-off between privacy and statistical accuracy.
She is co-author of “Half-Trek Criterion for Generic Identifiability of Linear Structural Equation Models,” and “Global Identifiability of Linear Structural Equation Models,” both of which were published in the Annals of Statistics.
She attended Brown University as an undergraduate and received an MS in mathematics and a PhD in statistics from the University of Chicago. During her final year of graduate study, she was awarded the Harper Dissertation-Year Fellowship. She was a National Science Foundation Mathematical Sciences Postdoctoral Research Fellow in the Statistics Department at Stanford University.
She joined the University of Chicago faculty in 2013.