• Robert L Wolpert and Katja Ickstadt

    Simulation of Lévy Random Fields


  • Abstract:

    An efficient recently developed method, the Inverse Lévy Measure (ILM) algorithm, is presented for drawing random samples from gamma, skewed stable and other nonnegative independent-increment random fields, which we call Lévy random fields. The method is useful for computing posterior distributions in nonparametric hierarchical Bayesian statistical analysis. Algorithms are illustrated through prototype implementations in S-plus.


  • Available in ps (716kb) or pdf (318kb) formats.
  • Authors' (Robert Wolpert, Katja Ickstadt) home pages
  • In Practical Nonparametric and Semiparametric Bayesian Statistics (Müller, Dey & Sinha, eds.), pp. 227-242.