作者: Søren Asmussen , Jens Ledet Jensen , Leonardo Rojas-Nandayapa
DOI: 10.1007/S11009-014-9430-7
关键词: Inverse Laplace transform 、 Moment-generating function 、 Characteristic function (probability theory) 、 Mathematics 、 Lambert W function 、 Laplace transform applied to differential equations 、 Mathematical analysis 、 Laplace's method 、 Two-sided Laplace transform 、 Laplace transform
摘要: Integral transforms of the lognormal distribution are great importance in statis- tics and probability, yet closed-form expressions do not exist. A wide variety methods have been employed to provide approximations, both analytical numerical. In this paper, we analyse a approximation � L(θ ) Laplace transform which is obtained via modified version Laplace's method. This approximation, given terms Lambert W (·) function, tractable enough for applications. We prove that asymptotically equivalent as θ →∞ . apply result construct reliable Monte Carlo estimator it be logarithmically efficient rare event sense