A factor analysis of long-term bond spreads is performed by decomposing int
ernational interest rate spreads into national and global factors. The fact
ors are latent, and are assumed to have GARCH-type specifications as well a
s exhibiting serial dependence. An indirect estimator is used to compute es
timates of the unknown parameters. The sampling performance of this estimat
or is investigated and compared with an alternative direct estimator based
on the Kalman predictor. The factor model is applied to weekly data on long
-bond spreads between five countries - Australia, Japan, Germany, Canada an
d the UK - and the USA over the period 1991 to 1999. The resulting factor d
ecomposition is used to examine the international investor's optimal portfo
lio decision in a mean-variance framework. Copyright (C) 2000 John Wiley &
Sons, Ltd.