English

A geometric interpretation of the transition density of a symmetric L\'evy Process

Probability 2011-05-26 v1

Abstract

We study for a class of symmetric L\'evy processes with state space \rn\rn the transition density pt(x)p_t(x) in terms of two one-parameter families of metrics, (dt)t>0(d_t)_{t>0} and (δt)t>0(\delta_t)_{t>0}. The first family of metrics describes the diagonal term pt(0)p_t(0); it is induced by the characteristic exponent ψ\psi of the L\'evy process by dt(x,y)=tψ(xy)d_t(x,y)=\sqrt{t\psi(x-y)}. The second and new family of metrics δt\delta_t relates to tψ\sqrt{t\psi} through the formula exp(δt2(x,y))=\Ff[etψpt(0)](xy) \exp(-\delta_t^2(x,y)) = \Ff[\frac{e^{-t\psi}}{p_t(0)}](x-y) where \Ff\Ff denotes the Fourier transform. Thus we obtain the following "Gaussian" representation of the transition density: pt(x)=pt(0)eδt2(x,0)p_t(x)=p_t(0) e^{-\delta_t^2(x,0)} where pt(0)p_t(0) corresponds to a volume term related to tψ\sqrt{t\psi} and where an "exponential" decay is governed by δt2\delta_t^2. This gives a complete and new geometric, intrinsic interpretation of pt(x)p_t(x).

Keywords

Cite

@article{arxiv.1105.5016,
  title  = {A geometric interpretation of the transition density of a symmetric L\'evy Process},
  author = {N. Jacob and V. Knopova and S. Landwehr and R. L. Schilling},
  journal= {arXiv preprint arXiv:1105.5016},
  year   = {2011}
}