《Dynamical pricing of weather》.pdf

Q UANTITATIVE F I N A N C E V O L U M E 2 (2002) 189–198 RE S E A R C H PA P E R I N S T I T U T E O F P H Y S I C S P U B L I S H I N G quant.iop.org Dynamical pricing of weather derivatives 1 1,2 3 Dorje C Brody , Joanna Syroka and Mihail Zervos 1 Blackett Laboratory, Imperial College, London SW7 2BZ, UK 2 Centrica Plc., Charter Court, 50 Windsor Road, Slough, Berkshire, SL1 2HA, UK 3 Department of Mathematics, King’s College London, The Strand, London WC2R 2LS, UK E-mail: dorje@ic.ac.uk, j.syroka@ic.ac.uk and mihail.zervos@kcl.ac.uk Received 18 October 2001, in final form 28 February 2002 Published 30 May 2002 Online at stacks.iop.org/Quant/2/189 Abstract The dynamics of temperature can be modelled by means of a stochastic process known as fractional Brownian motion. Based on this empirical observation, we characterize temperature dynamics by a fractional Ornstein–Uhlenbeck process. This model is used to price two types of contingent claims: one based on heating and cooling degree days, and one based on cumulative temperature. We derive analytic expressions for the expected discounted payoffs of such derivatives, and discuss the dependence of the results on the fractionality of the temperature dynamics. 1. Introduction degree day index’, or HDD for short, defined in section 4. Recently, however, contracts written on other indices, such Weather derivatives are classic examples of incomplete as the cumulative temperature over a specified time period, ¨ markets (cf Bjork 1998, Brody 2000). The payoffs of have also attracted attent