Get access

Multifactor Index Variance: The Case of the SPX 2000 to 2010

Authors

  • Thaddeus Neururer,

    Corresponding author
    • Thaddeus Neururer is a Research Assistant at the Boston University, School of Management, Boston, Massacchusetts
    Search for more papers by this author
  • Andrew Kumiega

    1. Andrew Kumiega is a Quality Manager at the Infinium Capital Management, LLC, Adjunct Faculty, Stuart School of Business, Illinois Institute of Technology, Chicago, Illinois
    Search for more papers by this author

  • Thaddeus Neururer is grateful for the help and comments from Patricia Caffrey, Krish Menon, and Corey Neururer.

Correspondence author, Boston University, School of Management, 595 Commonwealth Avenue, Boston, MA 02215, e-mail: thadn@bu.edu

Abstract

Recent work in volatility modeling and options pricing has suggested that index variance may be best described by multifactor models. We utilize Independent Component Analysis to produce independent factors from the SPX sector ETF returns from 2000 to 2010. We find that the independent factors are quite different from the factors produced with Principal Components. After fitting asymmetric GARCH models to each of the factors and applying the sector weightings of the SPX we show that, on average, SPX variance is best modeled with two factors. In addition, we also find that the secondary factor of SPX variance has switched from being technology based to being financial in nature. © 2011 Wiley Periodicals, Inc. Jrl Fut Mark 33:158–182, 2013

Get access to the full text of this article

Ancillary