Universality at the edge for unitary matrix models (Q2836982)

From MaRDI portal





scientific article; zbMATH DE number 6186195
Language Label Description Also known as
default for all languages
No label defined
    English
    Universality at the edge for unitary matrix models
    scientific article; zbMATH DE number 6186195

      Statements

      0 references
      10 July 2013
      0 references
      unitary matrix models
      0 references
      local eigenvalue statistics
      0 references
      universality
      0 references
      polynomials on the unit circle
      0 references
      math-ph
      0 references
      math.MP
      0 references
      math.PR
      0 references
      Universality at the edge for unitary matrix models (English)
      0 references
      Unitary matrix models are studied. These models are defined by the probability law NEWLINE\[NEWLINEp_n(U)d \mu_n(U) = Z^{-1}_{n,2}\text{exp}\left \{-n\text{Tr}V \left(\frac{U + U^*}{2} \right)\right \} d\mu_n(U),NEWLINE\]NEWLINE where \(U = \{ U_{jk}\}^n_{j,k=1}\) is an \(n \times n\) unitary matrix, \(\mu_n(U)\) is the Haar measure on the group \(U(n)\), \(Z_{n,2}\) is a normalization constant, and \(V : [-1, 1] \rightarrow {\mathbb R}\) is a continuous function called the potential of the model. As the main result, the universality conjecture is proved concerning the local eigenvalue statistics for unitary matrix models with a smooth potential \(V\) having four bounded derivatives in the case of one-interval support \(\sigma\) of the limiting normalized counting measure. This result is obtained using the orthogonal polynomial technique when a system of polynomials orthogonal on the unit circle with a varying weight is considered.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references