The dispersion bias

LR Goldberg, A Papanicolaou, A Shkolnik - SIAM Journal on Financial …, 2022 - SIAM
We identify and correct excess dispersion in the leading eigenvector of a sample covariance
matrix when the number of variables vastly exceeds the number of observations. Our …

[PDF][PDF] Better betas

LR Goldberg, A Papanicolaou, A Shkolnik… - The Journal of …, 2020 - dornsife.usc.edu
Compare the accuracy of long-only minimum variance portfolio weights and risk forecasts
using the dispersion bias correction and standard models, including vanilla PCA, beta …

James–Stein for the leading eigenvector

LR Goldberg, AN Kercheval - Proceedings of the National …, 2023 - National Acad Sciences
Recent research identifies and corrects bias, such as excess dispersion, in the leading
sample eigenvector of a factor-based covariance matrix estimated from a high-dimension …

[PDF][PDF] Portfolio optimization via strategy-specific eigenvector shrinkage

LR Goldberg, H Gurdogan, A Kercheval - 2023 - math.fsu.edu
We estimate covariance matrices for portfolio optimization that are tailored to given
constraints. Our factor-based, data-driven construction relies on a generalized version of …

Identifying broad and narrow financial risk factors with convex optimization

A Shkolnik, LR Goldberg, J Bohn - Available at SSRN 2800237, 2016 - papers.ssrn.com
Factor analysis of security returns aims to decompose a return covariance matrix into
systematic and specific risk components. To date, most commercially successful factor …

Equally weighted cardinality constrained portfolio selection via factor models

JF Monge - Optimization Letters, 2020 - Springer
In this work a proposal and discussion of two different 0-1 optimization models is carried out
in order to solve the cardinality constrained portfolio problem by using factor models. Factor …

Analysis of James-Stein for the Leading Eigenvector

S Ribas - 2023 - diginole.lib.fsu.edu
We examine the results of” James Stein for the leading eigenvector”(JSE)[6], which corrects
excess dispersion in the leading eigenvector of a factor-based covariance matrix estimated …