Hankel operators induced by radial Bekollé-Bonami weights on Bergman spaces (Q2197644): Difference between revisions

The authors investigate when big Hankel operators, which are defined by the formula \[ H^{\nu}_{f}(g)(z)=(I-P_{\nu})(fg)(z),\ f\in L^{1}_{\nu},\ z\in \mathbb{D}, \] map boundedly the weighted Bergman space \(A_{\omega}^{p}\) into \(L^{q}_{\nu}\). They give a necessary and sufficient condition for natural classes of radial weights. The condition says that \(f\) should belong to an appropriately defined \(\mathrm{BMO}\)-type space. We shall describe the main result in detail and explain the notation. The case of standard weights (see the explanation below) was studied in [\textit{J. Pau} et al., Indiana Univ. Math. J. 65, No. 5, 1639--1673 (2016; Zbl 1448.47042)]. For a radial weight \(\nu\), the symbol \(L^{q}_{\nu}\) stands for the space of all (equivalence classes) of complex-valued measurable functions \(f\) in \(\mathbb{D}\) such that \[ \|f\|_{L_{\nu}}^{q}=\Bigl{(}\int_{\mathbb{D}}|f(z)|^{q}\nu(z)\,dA(z)\Bigr{)}^{1/q}<\infty. \] For a radial \(\omega\), the weighted Bergman space \(A_{\omega}^{p}\) is the subspace of \(L_{\omega}^{p}\) consisting of all holomorphic functions in \(L_{\omega}^{p}\). The symbol \(P_{\nu}\) stands for the Bergman projection induced by the weight \(\nu\), \[ P_{\nu}(f)(z)=\int_{\mathbb{D}}f(\zeta)\overline{B_{z}^{\nu}(\zeta)}\nu(\zeta)\,dA(\zeta),\ z\in \mathbb{D}, \] where \(B_{z}^{\nu}\) are the reproducing kernels of \(A_{\nu}^{2}\). When \(\nu\) is a radial weight, then \(A^{2}_{\nu}\) is a closed subspace of \(L^{2}_{\nu}\). The weight \(\nu\) belongs to the class of radial weights in \(B_{q}=\bigcup_{\eta>-1}B_{q}(\eta)\). Here, \(B_{q}(\eta)\) denotes the class of all weights \(\omega\) such that the Bergman projection \(P_{\eta}\) induced by the standard weight \((\eta+1)(1-|z|^{2})^{\eta}\) is bounded on \(L^{q}_{\omega}\), \(1<q<\infty\). These classes of weights were described by \textit{D. Bekollé} [Stud. Math. 71, 305--323 (1982; Zbl 0516.47016); \textit{D. Bekolle} and \textit{A. Bonami} [C. R. Acad. Sci., Paris, Sér. A 286, 775--778 (1978; Zbl 0398.30006)]. The main result of the paper (Theorem 1) says that the operators \(H_{f}^{\nu},H_{\bar{f}}^{\nu}\) with \(f\in L_{\nu}^{q}\) map boundedly the space \(A_{\omega}^{p}\) into \(L_{\nu}^{q}\), \(1<p\leq q<\infty\), if and only if \(f\in \mathrm{BMO}(\Delta)_{\omega,\nu,p,q}\). The weight \(\nu\) is assumed (as above) to be radial and belong to \(B_{q}\). The weight \(\omega\) belongs to the family \(\mathcal{D}:=\hat{\mathcal{D}}\cap \check{\mathcal{D}}\). A radial weight \(\omega\) belongs to \(\hat{\mathcal{D}}\) is there is a constant \(C=C(\omega)>1\) such that \[ \hat{\omega}(r)\leq C\hat{\omega}\Bigl{(}\frac{1+r}{2}\Bigr{)},\ 0\leq r<1, \] where \(\hat{\omega}(z)=\int_{|z|}^{1}\omega(s)\,ds\) (this is assumed \(>0\)). If there exist \(K=K(\omega)>1\) and \(C=C(\omega)>1\) such that \[ \hat{\omega}(r)\geq C\hat{\omega}\Bigl{(}1-\frac{1-r}{K}\Bigr{)},\ 0\leq r<1, \] then \(\omega \in \mathcal{D}\). The space \(\mathrm{BMO}(\Delta)_{\omega,\nu,p,q,r}\) consists of \(f\in L^{q}_{\nu,\mathrm{loc}}\) such that \[ \|f\|_{\mathrm{BMO}(\Delta)_{\omega,\nu,p,q,r}}=\sup_{z\in \mathbb{D}} MO_{v,q,r}(f)(z)\gamma(z)<\infty. \] For \(0<p,q<\infty\) and radial weights \(\omega,\nu\), the function \(\gamma\) is defined as \[ \gamma(z)=\gamma_{\omega,\nu,p,q}(z)=\frac{\hat{\nu}(z)^{1/q}(1-|z|)^{1/q}}{\hat{\omega}(z)^{1/p}(1-|z|)^{1/p}}, \ z\in \mathbb{D}. \] Also, for \(f\in L^{1}_{\nu,\mathrm{loc}}\), \[ MO_{\nu,q,r}(f)(z):=\Bigl{(}\frac{1}{\nu(\Delta(z,r))}\int_{\Delta(z,r)}|f(\zeta)-\hat{f}_{r,\nu}(z)|^{q}\nu(\zeta)\,dA(\zeta)\Bigr{)}^{1/q} \] and \[ \hat{f}_{r,\nu}(z)=\frac{1}{\nu(\Delta(z,r))}\int_{\Delta(z,r)}f(\zeta)\nu(\zeta)\,dA(\zeta). \] The symbol \(\Delta(z,r)\) stands for the hyperbolic disc of center \(z\) and radius \(r>0\). If \(\nu \in \mathcal{D}\), then \(\mathrm{BMO}(\Delta)_{\omega,\nu,p,q,r}\) does not depend on \(r\) for \(r\geq r_{0}\). Theorem 2 in the paper says that, under the same assumptions that \(1<p\leq q<\infty\), \(\omega\in \mathcal{D}\), and \(\nu \in B_{q}\) is radial, the big Hankel \(H_{\bar{f}}^{\nu}\) with coanalytic symbol \(\bar{f}\) for \(f\in A_{\nu}^{1}\) maps boundedly \(A_{\omega}^{p}\) into \(L^{q}_{\nu}\) if and only if \[ \|f\|_{\mathcal{B}_{d\gamma}}:=\sup_{z\in \mathbb{D}}|f^{'}(z)|(1-|z|)\gamma(z)+|f(0)|<\infty. \]