Gaussian analytic functions and operator symbols of Dirichlet type (Q783215)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Gaussian analytic functions and operator symbols of Dirichlet type
scientific article

    Statements

    Gaussian analytic functions and operator symbols of Dirichlet type (English)
    0 references
    0 references
    0 references
    11 August 2020
    0 references
    The authors obtain an important estimate which is interpreted in terms of Gaussian analytic functions in connection with the Dirichlet space. For the unit disk \(\mathbb D=\{z\in\mathbb C:|z|<1\}\), let \(A^2(\mathbb D)\) denote the subspace of \(L^2(\mathbb D)\) consisting of the holomorphic functions. The Dirichlet space is the space \(\mathcal D(\mathbb D)\) of analytic functions \(f\) with \(f'\in A^2(\mathbb D)\), equipped with the Dirichlet inner product \(\langle f',g'\rangle_{\mathbb D}:=\int_{\mathbb D}f'(z)\overline{g'}(z)\frac{dxdy}{\pi}\), \(z=x+iy\). Denote \(\mathcal D_0(\mathbb D):=\{f\in\mathcal D(\mathbb D):f(0)=0\}\). Form a jointly Gaussian pair \((\Phi(z),\Psi(z))\) of \(\mathcal D_0\)-Gaussian analytic functions \[\Phi(z):=\sum_{j=1}^{\infty}\frac{\alpha_j}{\sqrt j}z^j,\;\;\;\Psi(z):=\sum_{j=1}^{\infty}\frac{\beta_j}{\sqrt j}z^j,\] where the \(\alpha_j,\beta_j\) are independently distributed standard Gaussian variables, taken from a Gaussian Hilbert space, \(\beta_j\) from \(N_{\mathbb C}(0,1)\). The authors prove the following precise estimate. Theorem 1.5.3. For all jointly Gaussian processes \((\Phi,\Psi)\) consisting of \(\mathcal D_0\)-Gaussian analytic functions, we have the estimate \[\frac{1}{2\pi}\int_{\partial\mathbb D}|\mathbb E\Phi(r\zeta)\Psi(r\zeta)|^2|d\zeta|\leq2r^2\log\frac{e}{1-r^2}.\] Corollary 1.6.1. Let \(M=\{m_{j,k}\}_{j,k=1}^{\infty}\) be a complex-valued matrix which contracts the \(\ell^2\) norm. Then, for \(0\leq s<1\), we have the estimate \[\sum_{l=2}^{\infty}s^l\left|\sum_{j+k=l}(jk)^{-\frac{1}{2}}m_{j,k}\right|^2\leq2s\log\frac{e}{1-s}.\] For a bounded \(\mathbb C\)-linear operator \(\mathbf{T}\) on \(L^2(\mathbb D)\), the authors define the Dirichlet symbol as \[\mathcal P[\mathbf{T}](z,w):=\langle\mathbf{T}(\overline s_z),s_w\rangle_{\mathbb D},\;\;\;z,w\in\mathbb D,\;\;\;s_z(\zeta)=\frac{1}{1-\overline z\zeta}.\] The next theorem characterizes \(\mathbb E\Phi(z)\Psi(z)\) as the Dirichlet symbols associated with contractions on \(L^2(\mathbb D)\). Theorem 1.8.3. Given a pair of jointly Gaussian \(\mathcal D_0\)-Gaussian analytic functions \((\Phi(z),\Psi(z))\), there exists a norm contraction \(\mathbf{T}:L^2(\mathbb D)\to L^2(\mathbb D)\) such that \(\mathbb E\Phi(z)\Psi(w)=zw\langle\mathbf{T}\overline s_z,s_w\rangle_{\mathbb D}\), \(z,w\in\mathbb D\). On the other hand, given a norm contraction \(\mathbf{T}\) on \(L^2(\mathbb D)\), there exists a pair of jointly Gaussian \(\mathcal D_0\)-Gaussian analytic functions \((\Phi(z),\Psi(z))\) such that the above equality holds. After studying the correlations \(\mathbb E\Phi(z)\Psi(z)\), the authors widen the standard Bloch space to the mock-Bloch space (or Blochish space). Their bounds are interpreted in terms of McMullen's asymptotic variance. Also, the authors show how the correlations \(\mathbb E\Phi(z)\Psi(w)\) may be expressed as Dirichlet symbols of contractions on \(L^2(\mathbb D)\). The Dirichlet symbols of Grunsky operators are characterized in terms of a nonlinear wave equation.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    Gaussian analytic function
    0 references
    Dirichlet space
    0 references
    asymptotic variance
    0 references
    Grunsky operator
    0 references
    0 references
    0 references