Extracting factors from heteroskedastic asset returns

This paper proposes an alternative to the asymptotic principal components procedure of Connor and Korajczyk (J. Financial Econom. 15 (1986) 373) that is robust to time series heteroskedasticity in the factor model residuals. The new method is simple to use and requires no assumptions stronger than those made by Connor and Korajczyk. It is demonstrated through simulations and analysis of actual stock market data that allowing heteroskedasticity sometimes improves the quality of the extracted factors quite dramatically. Over the period from 1989 to 1993, for example, a single factor extracted using the Connor and Korajczyk method explains only 8.2% of the variation of the CRSP value-weighted index, while the factor extracted allowing heteroskedasticity explains 57.3%. Accounting for heteroskedasticity is also important for tests of the APT, with p-values sometimes depending strongly on the factor extraction method used. r 2001 Elsevier Science S.A. All rights reserved. JEL classification: G12; C13