Transition probabilities for degenerate diffusions arising in population genetics (Q1729702)

From MaRDI portal





scientific article
Language Label Description Also known as
default for all languages
No label defined
    English
    Transition probabilities for degenerate diffusions arising in population genetics
    scientific article

      Statements

      Transition probabilities for degenerate diffusions arising in population genetics (English)
      0 references
      0 references
      0 references
      28 February 2019
      0 references
      Generalized Kimura operators were introduced by \textit{C. E. Epstein} and \textit{R. Mazzeo} [Degenerate diffusion operators arising in population biology. Princeton, NJ: Princeton University Press (2013; Zbl 1309.47001)]. The aim of this work is the study of the stochastic processes associated to these operators. They are diffusion processes on manifolds with corners. The diffusion coefficient is strongly elliptic in the interior, but degenerates on the boundary; moreover the drift is pointing inward on the boundary, and some other technical assumptions are made. \par Under these assumptions, the authors prove that the martingale problem is well posed, and that the resulting process is a strong Markov process. Then they study the behaviour on the boundary which is a stratified space made of tangent boundary components with different dimensions. When the process hits such a component, it hits the interior and remains there, performing a diffusion on a lower dimensional manifold with corners. \par All details are clearly explained in a long introduction, which also contains discussion about previous literature.
      0 references
      0 references
      degenerate elliptic operators
      0 references
      compact manifold with corners
      0 references
      fundamental solution
      0 references
      Dirichlet heat kernel
      0 references
      caloric measure
      0 references
      Markov processes
      0 references
      transition probabilities
      0 references
      hitting distributions
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references

      Identifiers