LQG 2023/09/11 – 18:30 – James–Stein for the leading eigenvector – Lisa Goldberg – BlackRock

Special additional seminar – James–Stein for the leading eigenvector

Seminar by – Lisa Goldberg

11th September 2023 – In-Person and On-Line – Hybrid

The LQG is very grateful to BlackRock for hosting this additional seminar.

In addition to speaking to the autumn seminar in Cambridge Lisa Goldberg will deliver this additional seminar on : James–Stein for the leading eigenvector while she is in London.
Note that 2023/09/11 is a MONDAY and not in the usual LQG sequence – this is a special opportunity.
The background material for this seminar is available here.  / https://www.pnas.org/doi/10.1073/pnas.2207046120
The latest material on which this seminar will be based is accessible here. / https://cdar.berkeley.edu/publications/portfolio-optimization-strategy-specific-eigenvector-shrinkage

Eigenvectors are used throughout the physical and social sciences to reduce the dimension of complex problems to manageable levels and to distinguish signal from noise. Our research identifies and mitigates bias in the leading eigenvector of a sample factor-based covariance matrix estimated in the high-dimension low sample size (HL) regime. The analysis illuminates how estimation error in a covariance matrix can affect quadratic optimization. Eigenvector estimation in the HL regime may be useful for disciplines, such as finance, machine learning, or genomics, in which high-dimensional variables need to be analyzed from a limited number of observations.


Lisa Goldberg

Managing Director, Head of Research at Aperio Group by BlackRock. Professor of the Practice of Economics and co-Director of the Consortium for Data Analytics in Risk, University of California, Berkeley.

Lisa Goldberg is Head of Research at Aperio Group, now part of BlackRock, and Professor of the Practice of Economics and co-Director of the Consortium for Data Analytics in Risk at University of California, Berkeley. She has worked in topology, dynamical systems, quantitative finance, sports statistics and causal inference. She has published more than 60 articles and is co-author of Portfolio Risk Management, which was published by Princeton University Press in 2010. Lisa is inventor on five patents and serves on editorial boards of four quantitative finance journals and two book series. She is a member of the Advisory Council for the Museum of Mathematics, an arXiv moderator and an expert judge for the Moskowitz Prize for Socially Responsible Investing. Lisa is 2/3 of the way to her lifetime goal of swimming a lap around the equator.