Mathematical Finance & Financial Data Science Seminar

---

The Dispersion Bias

Speaker: Lisa Goldberg, University of California, Berkeley and Aperio Group

Location: Warren Weaver Hall 1302

Date: Tuesday, March 27, 2018, 5:30 p.m.

Synopsis:

Estimation error has plagued quantitative finance since Harry Markowitz launched modern portfolio theory in 1952.  Using random matrix theory, we characterize a source of bias in the sample eigenvectors of financial covariance matrices.  Unchecked, the bias distorts weights of minimum variance portfolios and leads to risk forecasts that are severely biased downward.  To address these issues, we develop an eigenvector bias correction.  Our approach is distinct from the regularization and eigenvalue shrinkage methods found in the literature.   We provide theoretical guarantees on the improvement our correction provides as well as estimation methods for computing the optimal correction from data.  This work is in collaboration with Alex Papanicolaou and Alex Shkolnik.

Bio:

Lisa Goldberg is co-Director of the Consortium for Data Analytics in Risk at University of California, Berkeley, where she is Adjunct Professor of Economics and Statistics.  She is also Director of Research at Aperio Group.  A mathematician by training, Lisa held positions at City University of New York, Mathematical Sciences Research Institute and the Institute for Advanced Study before moving to MSCI Barra in 1992.  She is a researcher with five patents and more than 50 publications in topology, complex dynamical systems and quantitative finance. Her current interests include applications of data science to financial economics, impact investing and basketball.

 

 

Notes: