Design of Eigenportfolios for US Equities Using Exponential Correlation Model

Ali N. Akansu, Anqi Xiong

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The eigen decomposition of Toeplitz matrix, with exponential correlations as its elements, to model empirical correlations of US equity returns is investigated. Closed form expressions for eigenvalues and eigenvectors of such Toeplitz matrix are available. Those eigenvectors are used to design the eigenportfolios of the model. The Sharpe ratios and PNL curves of eigenportfolios for stocks in Dow Jones Industrial Average (DJIA) index for the period from July 1999 to Nov. 2018 are calculated to validate the model. The proposed method provides eigenportfolios that closely mimic the eigenportfolios designed based on empirical correlation matrix generated from market data. The modeling of empirical correlation matrix brings new insights to design and evaluate eigenportolios for US equities and other asset classes.

Original languageEnglish (US)
Title of host publication2019 53rd Annual Conference on Information Sciences and Systems, CISS 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728111513
DOIs
StatePublished - Apr 16 2019
Event53rd Annual Conference on Information Sciences and Systems, CISS 2019 - Baltimore, United States
Duration: Mar 20 2019Mar 22 2019

Publication series

Name2019 53rd Annual Conference on Information Sciences and Systems, CISS 2019

Conference

Conference53rd Annual Conference on Information Sciences and Systems, CISS 2019
CountryUnited States
CityBaltimore
Period3/20/193/22/19

All Science Journal Classification (ASJC) codes

  • Information Systems

Keywords

  • Exponential correlation model
  • Karhunen-Loeve Transform
  • PNL curve
  • Sharpe ratio
  • Toeplitz matrix
  • eigen decomposition
  • eigenportfolios
  • principal component analysis

Fingerprint Dive into the research topics of 'Design of Eigenportfolios for US Equities Using Exponential Correlation Model'. Together they form a unique fingerprint.

Cite this