A Pearson Random Walk with Steps of Uniform Orientation and Dirichlet Distributed Lengths - Université de Rennes Accéder directement au contenu
Article Dans Une Revue Journal of Statistical Physics Année : 2010

A Pearson Random Walk with Steps of Uniform Orientation and Dirichlet Distributed Lengths

Gérard Le Caër

Résumé

A constrained diffusive random walk of n steps in ℝ d and a random flight in ℝ d , which are equivalent, were investigated independently in recent papers (J. Stat. Phys. 127:813, 2007; J. Theor. Probab. 20:769, 2007, and J. Stat. Phys. 131:1039, 2008). The n steps of the walk are independent and identically distributed random vectors of exponential length and uniform orientation. Conditioned on the sum of their lengths being equal to a given value l, closed-form expressions for the distribution of the endpoint of the walk were obtained altogether for any n for d=1,2,4. Uniform distributions of the endpoint inside a ball of radius l were evidenced for a walk of three steps in 2D and of two steps in 4D. The previous walk is generalized by considering step lengths which have independent and identical gamma distributions with a shape parameter q>0. Given the total walk length being equal to 1, the step lengths have a Dirichlet distribution whose parameters are all equal to q. The walk and the flight above correspond to q=1. Simple analytical expressions are obtained for any d≥2 and n≥2 for the endpoint distributions of two families of walks whose q are integers or half-integers which depend solely on d. These endpoint distributions have a simple geometrical interpretation. Expressed for a two-step planar walk whose q=1, it means that the distribution of the endpoint on a disc of radius 1 is identical to the distribution of the projection on the disc of a point M uniformly distributed over the surface of the 3D unit sphere. Five additional walks, with a uniform distribution of the endpoint in the inside of a ball, are found from known finite integrals of products of powers and Bessel functions of the first kind. They include four different walks in ℝ3, two of two steps and two of three steps, and one walk of two steps in ℝ4. Pearson-Liouville random walks, obtained by distributing the total lengths of the previous Pearson-Dirichlet walks according to some specified probability law are finally discussed. Examples of unconstrained random walks, whose step lengths are gamma distributed, are more particularly considered.

Dates et versions

hal-00658342 , version 1 (10-01-2012)

Identifiants

Citer

Gérard Le Caër. A Pearson Random Walk with Steps of Uniform Orientation and Dirichlet Distributed Lengths. Journal of Statistical Physics, 2010, 140 (4), pp.728-751. ⟨10.1007/S10955-010-0015-8⟩. ⟨hal-00658342⟩
282 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More