Working Papers 2002 – Abstracts
A Class of Nonlinear Stochastic Volatility Models and Its Implications
on Pricing Currency Options
Jun Yu, Zhenlin Yang and Xibin Zhang
This paper proposes a class of stochastic volatility (SV) models which
offers an alternative to the one introduced in Andersen (1994). The class
encompasses all standard SV models that have appeared in the literature,
including the well known lognormal model, and allows us to empirically
test all standard specifications in a convenient way. We develop a likelihood-based
technique for analyzing the class. Daily dollar/pound exchange rate data
reject all the standard models and suggest evidence of nonlinear SV. An
efficient algorithm is proposed to study the implications of this nonlinear
SV on pricing currency options and it is found that the lognormal model
overprices options.
Keywords: Box-Cox transformations, Stochastic volatility, MCMC,
Exchange rate volatility, Option pricing.
Next Abstract
