AN EMPIRICAL APPLICATION OF STOCHASTIC VOLATILITY MODELS

This paper studies the empirical performance of stochastic volatility models for twenty years of weekly exchange rate data for four major cu rrencies. We concentrate on the effects of the distribution of the exc hange rate innovations for both parameter estimates and for estimates of the latent volatility series. The density of the log of squared exc hange rate innovations is modelled as a flexible mixture of normals. W e use three different estimation techniques: quasi-maximum likelihood, simulated EM, and a Bayesian procedure. The estimated models are appl ied for pricing currency options. The major findings of the paper are that: (1) explicitly incorporating fat-tailed innovations increases th e estimates of the persistence of volatility dynamics; (2) the estimat ion error of the volatility time series is very large; (3) this in tur n causes standard errors on calculated option prices to be so large th at these prices are rarely significantly different from a model with c onstant volatility. (C) 1998 John Wiley & Sons, Ltd.