Premium
Multifactor Index Variance: The Case of the SPX 2000 to 2010
Author(s) -
Neururer Thaddeus,
Kumiega Andrew
Publication year - 2013
Publication title -
journal of futures markets
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.88
H-Index - 55
eISSN - 1096-9934
pISSN - 0270-7314
DOI - 10.1002/fut.20552
Subject(s) - variance (accounting) , volatility (finance) , index (typography) , econometrics , realized variance , principal component analysis , autoregressive conditional heteroskedasticity , variance components , factor analysis , economics , statistics , mathematics , computer science , accounting , world wide web
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