Methods for evaluating density functions of exponential functionals represented as integrals of geometric Brownian motion (Q2583513): Difference between revisions

The aim of this paper is to propose a novel approach on numerical computation of probability densities in \textit{M. Yor}'s formula [Adv. Appl. Probab. 24, No.~3, 509--531 (1992; Zbl 0765.60084)], which is used to represent exponential functionals as integrals in time of geometric Brownian motions, with applications in the theory of mathematical finance, diffusion processes in random media, and in stochastic analysis for hyperbolic spaces. Since Yor's formula is somehow complicate, the author considers the expressions of density as derived by \textit{D. Dufresne} [ibid. 33, No.~1, 223--241 (2001; Zbl 0980.60103), and shows the coincidence of the Yor's and Dufresne's formulas by numerical computations, for the values of the constant drift \(\mu= 0\) and \(\mu = 1\). As an application of Dufresne-based numerical computation for densities in the involved exponential functionals, the paper obtains a novel method to compute the price of an Asian option in the Black-Scholes model. For the numerical computations, the software package of NetNUMPAC (http:/\!/netnumpac.fuis.fukui\_u.ac.jp/) was utilized.